General Issue

Journal of the British Interplanetary Society

VOLUME 72 NO.5 MAY 2019 General issue

A REACTION DRIVE Powered by External Dynamic Pressure Jeffrey K. Greason DOCKING WITH ROTATING SPACE SYSTEMS Mark Hempsell MARTIAN DUST: the Formation, Composition, Toxicology, Astronaut Exposure Health Risks and Measures to Mitigate Exposure John Cain THE CONCEPT OF A LOW COST SPACE MISSION to Measure “Big G” Roger Longstaff

www.bis-space.com

ISSN 0007-084X PUBLICATION DATE: 26 JULY 2019 Submitting papers International Advisory Board to JBIS

JBIS welcomes the submission of technical Rachel Armstrong, Newcastle University, UK papers for publication dealing with technical Peter Bainum, Howard University, USA reviews, research, technology and engineering in astronautics and related fields. Stephen Baxter, Science & Science Fiction Writer, UK James Benford, Microwave Sciences, California, USA Text should be: James Biggs, The University of Strathclyde, UK ■ As concise as the content allows – typically 5,000 to 6,000 words. Shorter papers (Technical Notes) Anu Bowman, Foundation for Enterprise Development, California, USA will also be considered; longer papers will only Gerald Cleaver, Baylor University, USA be considered in exceptional circumstances – for Charles Cockell, University of Edinburgh, UK example, in the case of a major subject review. Ian A. Crawford, Birkbeck College London, UK ■ Source references should be inserted in the text in square brackets – [1] – and then listed at the Adam Crowl, Icarus Interstellar, Australia end of the paper. Eric W. Davis, Institute for Advanced Studies at Austin, USA ■ Illustration references should be cited in Kathryn Denning, York University, Toronto, Canada numerical order in the text; those not cited in the Martyn Fogg, Probability Research Group, UK text risk omission. Raghavan Gopalaswami, Aerospace Researcher, India ■ Captions must be labelled with their Fig. number and should be as short as possible. Lamartine Guimarães, Institute for Advanced Studies, Brazil Mark Hempsell, Hempsell Astronautics Ltd, UK Illustrations should be: Takuto Ishimatsu, Massachusetts Institute of Technology, USA ■ Colour or mono, but should be as close to print Les Johnson, Marshall Space Flight Center, USA resolution (300 dpi) as possible. Poor-quality illustrations may compromise the acceptance of Terry Kammash, University of Michigan, USA paper for publication. Images embedded in Word Kelvin F. Long, Initiative for Interstellar Studies documents may be acceptable, but JBIS reserves Inoue Makoto, Institute of Astronomy & Astrophysics Academia Sinica, Taiwan the right to request separate higher-resolution Gregory L. Matloff, City University New York, USA image files from the author prior to publication. Koichi Mori, Nagoya University, Japan ■ Responsibility for copyright clearance of images rests entirely with the author. Richard Obousy, Richard Obousy Consulting LLC, USA Robert Parkinson, BIS, Aylesbury, UK Submission of papers George Schmidt, NASA John H Glenn Research Center, Ohio, USA ■ Papers for consideration should be sent by Paul Schuch, The SETI League Inc, USA email to [email protected] as both a Word document and as a Word PDF file (in order to Tabitha Smith, Bifrost, USA check for font anomalies), together with any Andreas Tziolas, Variance Dynamical Corporation, USA separate image files. Chris Welch, The International Space University, Strasbourg, France ■ If a paper is accepted for publication, the Friedwardt Winterberg, University of Nevada, Reno, USA author will be asked to sign a License to Publish form. This can be downloaded at www.bis- space.com/wp-content/uploads/2012/08/ WebsiteLicense.pdf. ■ Authors will receive a complimentary copy of the issue in which their paper appears. Editor Roger Longstaff Deputy Editor Duncan Law-Green Associate Editors Stephen Ashworth, We respectfully ask authors to adhere to these Keith Cooper, Stephen Gamble, Paul Gilster, Rob Swinney, Production MP3 Media guidelines. Failure to do so will result in the Promotion Gill Norman JBIS Office British Interplanetary Society, Arthur C. Clarke House, delay of acceptable papers for publication. 27-29 South Lambeth Road, London, SW8 1SZ, United Kingdom tel +44 (0)20 7735 3160 email [email protected] www.bis-space.com

Our full Guidelines for Authors can be downloaded DISTRIBUTION from www.bis-space.com JBIS is distributed worldwide by mail and may be received by annual subscription or purchase of single copies. It is available through membership of the British Interplanetary Society at much reduced rates. Subscription details for members, non-members and libraries are available from the above address.

JBIS is a publication that promotes the mission of the British Interplanetary Society. Opinions expressed in signed articles are those of the contributors and do not necessarily reflect the views of the Editor or the Council of the British Interplanetary Society. Security clearance, where necessary, is the responsibility of the author.

Published by the British Interplanetary Society. Registered Company No: 402498. Registered Charity No: 250556. Printed by Latimer Trend & Company Ltd, Estover Road, Plymouth, PL6 7PY, England. © 2018 British Interplanetary Society. No part of this magazine may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying or recording by any information storage or retrieval system without prior permission from the Publishers. contents VOLUME 72 NO.5 MAY 2019

146 A REACTION DRIVE Powered by External Dynamic Pressure Jeffrey K. Greason

153 DOCKING WITH ROTATING SPACE SYSTEMS Mark Hempsell

161 MARTIAN DUST: the Formation, Composition, Toxicology, Astronaut Exposure Health Risks and Measures to Mitigate Exposure John Cain

172 THE CONCEPT OF A LOW COST SPACE MISSION to Measure “Big G” Roger Longstaff

OUR MISSION STATEMENT The British Interplanetary Society promotes the exploration and use of space for the benefit of humanity, connecting people to create, educate and inspire, and advance knowledge in all aspects of astronautics.

JBIS Vol 72 No.5 May 2019 145 JBIS VOLUME 71 2018 PAGES 146–152

A REACTION DRIVE Powered by External Dynamic Pressure

JEFFREY K. GREASON, Electric Sky, Inc., 602 North Baird St., Suite 200, Midland, TX, 79701, USA email [email protected]

A new class of reaction drive is discussed, in which reaction mass is expelled from a vehicle using power extracted from the relative motion of the vehicle and the surrounding medium, such as the solar wind. The physics of this type of drive are reviewed and shown to permit high velocity changes with modest mass ratio while conserving energy and momentum according to well-established physical principles. A comparison to past propulsion methods and propulsion classification studies suggests new mission possibilities for this type of drive. An example of how this principle might be embodied in hardware suggests accelerations sufficient for outer solar system missions, with shorter trip times and lower mass ratios than chemical rockets.

Keywords: Propulsion, Reaction drive, Solar wind, External power, Dynamic pressure

1 INTRODUCTION NOMENCLATURE In the sixty years since the first interplanetary spacecraft (Luna drag on the ship, in the ship frame (negative if it 1), scientific probes have been flown to all the large bodies in the decreases a positive velocity), N solar system, and, after decades of flight time, the twin Voyager specific impulse, m/s 1 and 2 spacecraft are entering the boundary between the solar power, (positive if supplied from ship, negative if system and interstellar space. However, missions to the outer supplied to the ship), W solar system are still very difficult, with long trip times, even thrust applied to ship (positive if it increases a with use of gravity assist maneuvers. positive velocity), N net thrust of the ship, surplus of thrust over drag, Substantial reductions in trip times to the outer solar system or for interstellar precursor missions are difficult for fun- N damental physical reasons. Fast trips imply high velocities: a reaction mass expelled, kg constant speed of 100 km/s is only ~20 AU/year, beyond any mass of the ship, kg demonstrated capability (though achievable with a close-solar mass of the surrounding medium which interacts flyby Oberth maneuver). Fast trips also imply that accelera- with the ship, kg tion cannot be too small: a 29 AU trip (Neptune from Earth) dynamic pressure of the medium on the ship, Pa of 100km/s peak velocity requires a constant acceleration of at velocity of the ship in the rest frame, m/s 2 least 0.005 m/s to achieve a two-year flight time (ignoring So- velocity of the surrounding medium in the rest lar gravity), otherwise too much time is spent in acceleration frame, m/s and braking to take advantage of high speed. before interaction with the propulsion system, in the rest frame, m/s With rocket propulsion, high velocity implies either high mass ratio (expense) or high exhaust velocity (high specific en- after interaction with the propulsion system, ergy of the propellant). High acceleration implies high specific in the rest frame, m/s power, which is why electric rockets have not been able to over- exh aust velocity of the reaction mass, relative to come these limitations. Nuclear propulsion systems offer high the ship, in the ship frame, m/s specific energy, but whether they can combine high specific freestream velocity; velocity of the surrounding energy with high specific power remains to be demonstrated. medium in the ship frame, m/s ρ density of the surrounding medium, kg/m3 These well-known challenges have led to exploration of various types of ‘sail’ which use either the photons or the solar wind particles as an external source of momentum to harvest vated by the realization that while such sails offer near-term [1,2]. Most of these approaches offer low accelerations because prospects for acceleration to high heliocentric velocities away of the large collection areas required, but at least one, the “plas- from the sun (hundreds of km/s), that there is no current pro- ma magnet” [3], offers useful accelerations by using large-scale pulsion system which permits braking from those velocities or magnetic fields from small generators. This work was moti- sunward acceleration.

146 Vol 72 No.5 May 2019 JBIS A REACTION DRIVE Powered by External Dynamic Pressure

The widely known methods of accelerating and decelerat- This simplified form illustrates that the higher the ing in a surrounding medium, including propellers, ramjets, freestream velocity, the more power is required for a giv- turbojets, rockets, parachutes, and sails, form distinct classes en thrust, which is why rockets tend to dominate at higher of propulsion. Energy can be provided by the vehicle or by speeds even when used within the atmosphere [8]. Recently, the surrounding medium, while reaction mass can be car- systems extending the propeller principle to the interplane- ried aboard or harvested from the surrounding medium. By tary plasma as a medium have been suggested, with the same classifying propulsion systems in this way (a “morphological general physical principles [9,10]. analysis”, following the methods of Zwicky [4]), a promising form of propulsion is identified, in which the reaction mass is 2.2 Rockets (internal energy, internal reaction mass) carried aboard the vehicle, but the energy to expel that reaction mass is provided by the passage of the vehicle through the As the limitations of propeller systems in reaching high veloc- medium. This was anticipated by Alan Bond [5] in the limit of ities became apparent, the application of the rocket principle high-speed operation of ram-augmented interstellar rockets in became attractive. All rocket-type systems, regardless of pow- which inert, rather than energetic, reaction mass could be used. er source, have broadly similar behavior. They are governed The principle however is useful in contexts beyond the original by the rocket equations [11]. application. We review the physics of the classical systems, and then explore the physics of this alternative form of propulsion. Finally, some examples of how this might be realized in an im- (6) plementable device for fast transportation in the interplanetary medium are given.

2 REVIEW OF CLASSICAL APPROACHES AND THEIR (7) PHYSICS

A review of the fundamental physics of existing propulsion is (8) needed to understand how this method differs. The methods are grouped depending on whether propulsive energy is internally carried or externally harvested, and whether reaction In rocket systems, the reaction mass that is ejected to con- mass is internally carried or externally harvested. Beginning serve momentum is carried aboard the ship, as is the energy with the equations of those well-known systems also provides that is converted into the kinetic energy of both the ship and the basis for deriving the physics of the new approach. the exhaust. For example, the chemical energy of fuels and oxidizers are converted to reaction mass. The rocket equation 2.1 Propeller Systems (internal energy, external reaction is derived from the conservation of energy and momentum. mass) While such a form of propulsion works in a vacuum, the amount of velocity gain is limited by the onboard energy and The earliest known forms of propulsion (rowing, paddlewheels, mass storage. In the case of chemical rockets, with practical propellers, turbojets, ramjets) involve pushing against the me- exhaust velocities of ≤4500 m/s, maximum vehicle propulsive dium surrounding the vehicle, using energy carried aboard the velocity gains of ~20000 m/s are the greatest achieved to date, vehicle. These forms of propulsion use the same basic phys- though missions with higher heliocentric velocities have been ics: they are reaction drives, accelerating the medium around achieved by gravity assist maneuvers. the vehicle [6]. They all depend on the surrounding medium, and the energy requirements to produce thrust increase with 2.3 Drag Devices (external energy, external reaction mass) speed relative to the medium, governed by the propeller equations [7], so maximum velocities are limited. Where is Where the surrounding medium is moving relative to the the streamtube of the surrounding medium captured by the ship, the application of drag can be useful, either to acceler- propulsion device, and to which mechanical work is done, and ate downwind (simple sails) or to brake a preexisting velocity ∆ is the change in velocity of the wind caused by the pro- (parachutes and aerobrakes). In these cases, any energy re- pulsion system: quired is provided by (or carried away by) the surrounding (1) medium, and the reaction mass is also formed by the surrounding medium. Drag devices are usually considered a distinct class of device from propellers and rockets. Where (2) One motivation for the current work is the recent prolif- eration of proposals for using the interplanetary or interstel- (3) lar plasma as a medium for drag devices, which show that in spite of the low density (~10-20 kg/m3 for the interplanetary medium at 1 AU, as low as ~10-22 kg/m3 in hot plasma inter- With defined by: (4) stellar regions), useful accelerations can be achieved through electromagnetic interactions. This was first conceived as a Note that care is required in observing the sign of these magnetic sail or magsail [1], and more recently as an elec- quantities, because the ship, medium, and reaction mass are tric sail [2]. A particularly high drag to mass configuration is all moving relative to each other. In the case of high freestream the “plasma magnet” magnetic sail, which offers a streamtube velocities ( << ), Equation 3 becomes: capture area far larger than the physical dimension of the coils involved in the device [3]. Fundamentally these are all (5) drag devices, although the capture area, and hence the value of involved at a given phase of flight, differ signifi-

JBIS Vol 72 No.5 May 2019 147 JEFFREY GREASON cantly. As drag devices, they provide thrust as in Equation 1 to operate a drive on these principles; one need only extract en- above, although the power, as shown in Equation 5, is then ergy from it to expel onboard inert reaction mass. delivered to the ship rather than being provided by the ship. Power can be large in cases where | | is large. 3.1 Nomenclature of This Type of Drive

These drag devices have great promise for certain missions The nomenclature for such a device is not obvious. While it including outer solar system flybys or missions to the helio- might be classified under the broad heading of ‘jet propulsion’ sphere boundary, and for braking systems for interstellar mis- since it expels reaction mass, that classification also includes sions. By harvesting thrust power from outside sources, they propellers, which are broadly recognized as different from can operate at levels of thrust power well beyond our current rockets. As will be seen, the governing equations are also dif- ability to provide propulsive energy storage aboard a space- ferent from rockets (the rocket equation does not apply), so craft. Unfortunately, by the nature of a drag device, they can calling them some form of ‘rocket’ seems misleading. And only accelerate “downwind”, and so can only partially reduce since they produce thrust and consume propellant mass, ‘sail’ propulsion requirements in cases such as outer solar system or- hardly seems appropriate. Following Zwicky, one might think biters. Many desirable missions require thrust both for acceler- of them as a ‘dynamic-pressure-powered mass driver’, but that ation and for deceleration (stopping and starting a fast transit). is rather clumsy. Bond [5] suggests this as the high-speed, inert reaction mass limit of a ram-augmented interstellar rocket, but It is worth noting that the ideal Bussard Ramjet [12] while since in the general implementation, there is neither ram-pres- not a ‘drag’ device, would also fall in to this category of both the sure recovery, nor a rocket, nor augmentation, nor interstellar energy and the reaction mass being provided externally. The flight, that nomenclature seems ill-suited to the general case. many practical difficulties in implementation of such a device This propulsive principle might be called a “wind drive”, or, have been discussed in the literature beginning with [13]. “ram drive”, but using the common abbreviation q for dynamic pressure [15] suggests the name q-drive – which is the name 3 THE REMAINING OPTION used in the balance of this text.

A morphological analysis (Zwicky box [4]) of the suite of pro- 3.2 Momentum and Energy Conservation pulsion devices shows that there is a remaining class of reaction devices: one in which the reaction mass is carried aboard the Fundamentally, as a propeller takes advantage of the fact that ship and is expelled using the power extracted from the flow at low speed, it takes little energy to make thrust, the q-drive of the surrounding medium. This approach does not appear in principle takes advantage of the fact that at high speed, a small the common surveys of the propulsion art [4,11,14], and the drag device can extract a great deal of power. The power from a first mention of it appears to be in Bond’s discussion of the wind-harvesting device follows Equation 5, while the power re- Ram-Augmented Interstellar Rocket [5], in which he points quired to expel stored reaction mass follows Equation 8. In the out that in the limit of high speed operation, the energy contri- ideal case of no losses and no parasitic drag, this leads to the bution of the rocket propellant becomes nearly negligible and following fundamental equations (derived in the Appendix) for that indeed the RAIR could then function with inert reaction minimum use of reaction mass: mass. However, there is no reason to limit the application of (9) this principle to that particular implementation – indeed, as noted in [5], the process of ram-compression of the interstellar medium to densities where RAIR operation is plausible intro- (10) duces inefficiencies (parasitic drag) which make that particular implementation difficult (Fig.1). (11) The key element of the type of drive contemplated here is that if the interplanetary or interstellar medium is dense enough to provide meaningful drag using plasma techniques, then it Contrast Equation 11 with Equation 6 and three dramatic dif- can be a source of power as well as drag. The medium can do ferences are apparent, all favoring the q-drive principle in high mechanical work on a system, thus extracting power from the velocity flight. First, in cases where is large compared ‘wind’: analogous to a ram air turbine in atmospheric flight. to a rocket exhaust velocity ( ), the scaling is more favorable Since in doing so the vehicle experiences drag, the fundamental for the q-drive. Second, mass ratio for a q-drive scales with the equations of this class of propulsion system must be examined square of velocity rather than with the exponential of veloc- to determine its performance and behavior. There is no need in ity as in a rocket. Third, in cases where ∆v is much less than general to compress the interstellar or interplanetary medium , as in most flight in the solar system due to the high

Fig.1 Morphological box of propulsion methods.

148 Vol 72 No.5 May 2019 JBIS A REACTION DRIVE Powered by External Dynamic Pressure velocity solar wind, the required mass ratio is even smaller 4 EXAMPLE IMPLEMENTATIONS (bearing in mind that the q-drive principle is only useful in situations where >>0). The propulsive principle outlined in this paper could apply to high-speed atmospheric flight or to travel in the interplanetary Two examples help to illustrate the q-drive principle. Con- or interstellar medium. However, to determine whether the sider operating in a medium that is essentially at rest in the q-drive principle has real engineering utility, some concept of stationary reference frame, such as the interstellar medium how this principle can be embodied in hardware is helpful. Fur- in heliocentric coordinates. If given (through the use of some thermore, the question of whether the acceleration achieved other propulsion system), an initial velocity of 600 km/s, is useful for fast transits can only be assessed in the light of a which for zero wind speed is also of 600km/s, using the hardware implementation. Realize that these examples are just q-drive principle with a mass ratio of 16 gives a final velocity that: guideposts for two ways to apply the q-drive principle to of 2400 km/s. This rather startling velocity does not rely on an real hardware. The first example is presented only because it is onboard nuclear reactor or energetic propellant; it is simply the physically very simple, and so enhances understanding of the result of momentum and energy exchange with the rest of the physical principles. The second example may be a practical im- medium. The reaction mass is carried away by the surrounding plementation, with acceleration > 0.02 m/s2. medium and is at rest with respect to it, so the kinetic energy of the initial high-mass ship plus reaction mass has been concen- 4.1 Continuous Mode, Electric Field Power Extraction trated into a final, low-mass ship at much higher velocity. It is worth noting that use of drag devices such as the Plasma Mag- Flow of the solar wind or interstellar plasma over electrodes net sail purely in a drag configuration can produce heliocentric can be used to generate electrical power to expel reaction mass, velocities of this magnitude, and that the abrupt deceleration of following the q-drive principle. Flow of a neutral plasma across the solar wind in the termination shock at the heliopause then a tandem pair of grids, with the solar wind flowing over them, means that same heliocentric velocity, which tended towards will develop a voltage difference from which power can be ex- of zero within the solar wind now presents a high in the tracted. This principle is well known as a means of extracting interstellar medium. power from conceptual fusion reactors [17], and its use in the reversed mode, applying power to make thrust, is noted in [9]. The second example is a case relevant to maneuvering inside Because the Debye sheath formed around each conductor lim- the solar system. Consider the solar wind to have a constant ve- its the amount of plasma intercepted, this approach requires locity of 450 km/s, and suppose a ship has been brought to a ve- high mass and offers low acceleration, but the principle of op- locity radially outward from the sun of 150km/s (for example, eration is helpful to understand. The velocity of the wind over by the use of a plasma magnet drag device). To brake from that the ship produces electrical power. Extracting that power outward velocity to achieve a state of rest in heliocentric coor- creates a voltage difference between the grids, which manifests dinates is then a ∆v of 150km/s, where the relative ‘wind’ speed as drag, precisely as in a windmill or ram air turbine operating is initially 300 km/s and rises during the maneuver to 450 in the air. Lower mass might be achieved by using a tandem set km/s. (When the vehicle is at rest in heliocentric coordinates, of radial wires similar to the “e-sail” [18]. it has equal to the wind speed.) In this case, the mass ratio required is 2.25 from Equation 11. By comparison, to achieve In turn, the electrical power can be used to expel reaction the same maneuver with the same mass ratio using a rocket, an mass. Any type of electrically powered thruster could be used, exhaust velocity of 185 km/s would be required, which is far provided the reaction mass can be expelled at approximately beyond any chemical rockets’ capability, and if based on an on- the same exhaust velocity as the freestream velocity (| | = ). board power plant, would require a very high power-to-mass While existing electric thrusters operating at ~4x105 m/s ex- ratio. By using the q-drive principle, the result can be achieved haust velocity are immature, they are plausible under known with inert reaction mass and with power harvested from the physical principles [19]. The expelled reaction mass ends up motion of the ship through the surrounding medium. nearly at rest with respect to the solar wind, while the ship ac- celerates sunward (or reduces its outward velocity). At first glance, the q-drive principle appears to offer “some- thing for nothing”. Propellant is expended but where does the 4.2 Pulsed Mode, Magnetic Field Power Extraction energy come from? The answer is that the energy comes from the loss of velocity of the reaction mass to the surrounding me- For accelerations that enable fast transits, a method of extract- dium. One may think of it as an inelastic collision between the ing power from the solar wind is needed that provides a high expended reaction mass and the surrounding medium, where drag-to-mass ratio, and it seems likely that a low parasitic drag the resulting change in energy is carried away by the ship. In is also important. In atmospheric applications, rotating devic- this sense, it is very reminiscent of the Oberth effect [16], in es (windmills, anemometers) are used to draw power from the which there are also three masses involved: the ship, the ex- wind, and magnetic field analogies of both are possible, but haust mass, and a planet. The q-drive principle is much more the relatively low lift-to-drag ratio of magnetic fields in plas- flexible, however, since it uses the surrounding medium as the ma suggests these approaches may have high parasitic drag. A third mass, and so the q-drive is not restricted to operation useful approach may lie in a linear, reciprocating motion of a near a gravitating body. magnetic field, where essentially all the drag goes into pushing on a moving field. High drag-to-mass is achievable using the Finally, while the analogy is imperfect, this has some sim- plasma magnet approach [3]. ilarity to the method by which sailing vessels on Earth can sail upwind. In that case, the energy is derived from the mo- The basic principle of the plasma magnet, illustrated in Fig. tion of the surrounding air, and the “reaction mass” is pro- 2 overleaf, is that a rotating magnetic field, driven by alter- vided by the action of the keel on the water. In space, we can nating current in a crossed pair of coils, creates a circulating achieve comparable results by expelling reaction mass from current in the plasma, and that current then expands in radi- the vehicle. us until it creates a dipolar magnetic field much larger than

JBIS Vol 72 No.5 May 2019 149 JEFFREY GREASON the physical coils.

If such a field is turned on and the generating coils are attached to a tether, the tether will be pulled by the solar wind, which could rotate the shaft of a conventional generator. Then, the field could be turned off, the tether reeled back in, and the cycle repeated. In principle this approach of mechan- ically moving the field coils in a reciprocating manner would extract power, and it illustrates the principle involved, but the mechanical motions would be too slow to provide adequate power-to-mass ratio. We need a more rapid motion of the field, which can be achieved by replacing the reciprocating motion of the coils carrying the magnetic field with the reciprocating motion of the magnetic field itself.

In the approach illustrated in Fig.3, a pair of plasma magnet generating coil sets are used, separated by a tether with Fig.2 Operating principle of a plasma magnet. wires to transfer power from one set of coils to the other. In- itially, the windward coil set is energized and the solar wind require that particular implementation, nor do they require pushes on it, transferring the energy in the dipole field to the fusion technology, and by exploiting the solar wind, they are leeward coils. During the power stroke, energy is extract- particularly useful for interplanetary flight. A conceptual de- ed from the wind, which can be used to power an electric sign suggests that, by using plasma magnet techniques, such thruster to expel reaction mass. A third coil set, omitted from a drive could offer accelerations and mass ratios sufficient for the illustration for clarity but located at the windward end rapid transits to the outer solar system. with a closed (toroidal) configuration that does not generate a magnetic field outside the coils, receives the energy on the To explore further, the analysis of the physics involved return stroke, so that drag is only pushing on the field during needs to be extended in two ways. First, the analysis needs the power stroke. Then, the energy is again transferred to the to include the effects of efficiencies in power conversion and windward coil, and the cycle repeats. parasitic drag, to assess whether the approach is practical. Second, to extend the application of this technique for inner A detailed design would be required to estimate mass but solar system missions, the theory needs to be extended to in- a sizing study, based on peak currents in superconducting clude thrusts that are not parallel to the drag vector, which 8 2 MgB2 tapes at 20K [20-22] of 2.5 x10 A/m , suggests that ac- would enable a wider range of maneuvers. celerations in the 0.025-0.05 m/s2 range may be feasible using this approach. The long tether, carrying oscillating currents This paper begins to examine routes for embodying this in the 1 KHz range from end to end, modulated by a recip- type of reaction drive in hardware. To assess the achievable rocating frequency in the 20 Hz range, is admirably suited to accelerations, designs will need to be carried to a level of de- form a Wideröe style [23] ion accelerator, thus providing an tail at which masses can be estimated credibly. integrated method for converting the resulting electric power to thrust. Acknowledgments

5 CONCLUSION While this work was developed without sponsorship from any organization, the author thanks the Tau Zero Foundation A new class of reaction drives appears capable of generating and the many researchers who have participated in the NASA vehicle velocities greater than those practical for propeller Institute for Advanced Concepts for their work in advanced or rocket devices. The basic principles of this drive are those propulsion, elements of which inspired the investigation employed in the “inert reaction mass, high velocity limit” leading to this work. Thanks also to the reviewers at JBIS who of the Ram-Augmented Interstellar Rocket, but they do not suggested important prior art to include.

Fig.3 Oscillating magnetic piston for energy extraction from solar wind.

150 Vol 72 No.5 May 2019 JBIS A REACTION DRIVE Powered by External Dynamic Pressure

APPENDIX A

The derivation of the equations for the q-drive principle begins The choice ofφ which minimizes reaction mass can be de- with, in the ship frame, recasting Equation 1 for the case where termined by maximizing specific impulse: the drag comes from a ‘windmill’ extracting power from the wind, and thrust comes from the expulsion of reaction mass, (A10) as in Equation 7:

(A1) By substituting Eq. (A2) in to Eq. (A10), and then in turn substituting Eq. (A8) we get:

(A2) (A11)

We can then define a net thrust , which, due to our sign Which has a maximum at φ = 0.5 producing the result, at convention, is the sum (not the difference) of thrust and drag. that φ = 0.5 condition, which will be assumed throughout the rest of the analysis: (A3) (A12) Knowing that energy and momentum are both conserved in all reference frames, we begin in the ship frame, where mo- Substituting back in the definition of ISP gives: mentum conservation gives rise to Eq. (A1) and (A2) while energy conservation gives us, using the same sign convention: (A13)

(A4) Observing that as long as is constant, ∆ =∆ and rearranging in terms of small changes in mass and velocity, and then realizing that every increment of exhaust mass is a decre- If , the equations are simplified, since ment in ship mass: in a drive powered by the dynamic pressure of the passing medium, ≈0 describes a condition where thrust drops to zero, (A14) these approximations are sound. Furthermore, for the ‘drive’ condition, in our sign convention, ∆ >0, as the acceleration The same result can be derived by energy conservation in from thrust, must be ‘into the wind’. If thrust were negative, it the rest frame, though the derivation is more complex. would represent a drag device, in which case no reaction mass or use of the q-drive principle would be needed. Drag power is To solve for the relationship between total mass expended negative because it is power supplied to the ship. That allows and velocity change, recognize that Eq. (A14) is a differential simplification of Eq. (A4) to: equation and rearrange as:

(A5) or, equivalently (A15)

Energy conservation for the thrust gives us: Which has a solution of the form = where C is a constant of integration, and a function of . The ratio of (A6) initial mass (before a maneuver) and final mass (after a maneuver) is then: Which can be rewritten as: (A16) (A7)

While in actual cases it may be desirable to use propellant Realizing that for any maneuver using the q-drive principle, that adds energy to the available thrust power, for simplicity in since is always a headwind (positive) and acceleration is al- the analysis consider the case where there is zero energy con- ways in the direction of increasing (if it were not, we would tent in the propellant, where it is inert reaction mass. Again, use a sail without need for expelling reaction mass), we can neglecting unavoidable inefficiencies for simplicity, that -im rewrite this as: plies that the power available for thrust is equal to the power derived from the drag, .

This produces the simple and useful result, using Eq. (A5) and (A7) as the thrust and drag power, that:

(A8)

The relationship between thrust and drag is a design param- eter; in a real system with less than perfect efficiencies it will be set by engineering considerations. In the ideal case, however, simply set a ‘thrust fraction’;

(A9)

JBIS Vol 72 No.5 May 2019 151 JEFFREY GREASON

REFERENCES 1. R. Zubrin and D. Andrews, “Magnetic Sails and Interplanetary Travel,” 12. R.W. Bussard, “Galactic Matter and Interstellar Flight”, Acta 25th AIAA/ASME/SAE/ASEE Joint Propulsion Conference, AIAA 89- Astronautica, 6, 1960, pp. 179-194. 2441, AIAA, Washington, DC, 1989. 13. T.A. Heppenheimer, “On the Infeasibility of Interstellar Ramjets”, JBIS, 2. P. Janhunen, “Electric Sail for Spacecraft Propulsion,” AIAA Journal of 31, 1978, p. 222ff Propulsion and Power, 20(4), 2004, pp 763-764. 14. R. Forward, “Alternate Propulsion Energy Sources,” Air Force Rocket 3. J. Slough and L. Giersch, “The Plasma Magnet”, 41st AIAA/ASME/ Propulsion Laboratory, AFRPL TR-83-067, December 1983. SAE/ASEE Joint Propulsion Conference, AIAA 2005-4461, AIAA, 15. M. Munk, “Notes on Propeller Design – IV”, NACA Technical Note No. Washington, DC 2005. 96, May, 1922, p. 6 4. F. Zwicky, “Fundamentals of Propulsive Power,” International 16. H. Oberth, Wege zur Raumschiffahrt, R. Oldenbourg Verlag, Munich- Conference for Applied Mechanics, Paris, September 1946. Collected in Berlin, 1929, in translation as “Ways to Spaceflight,” NASA TT F-622, Morphology of Propulsive Power, Society for Morphological Research, 1970, pp. 217-229 Pasadena, 1962, p. 37-47. 17. R. Hirsch, “Inertial Electrostatic Confinement of Ionized Fusion Gases”, 5. A. Bond, “An Analysis of the Potential Performance of the Ram Journal of Applied Physics, 38(11), 1967 Augmented Interstellar Rocket”, JBIS, 27, 1974, pp. 674-685. 18. B. Wiegmann, J. Vaughn, T. Schneider, M. Bangham, and A. Heaton, 6. W.J.M. Rankine, “On the Mechanical Principles of the Action of “Findings from NASA’s 2015-2017 Electric Sail Propulsion System Propellers,” Transactions of the Institution of Naval Architects, 6, 1865, Investigations”, 31st Annual AIAA/USU Conference on Small Satellites, pp. 13-39. AIAA, Washington, DC 2018. 7. W.E. Froude, “On the Part Played in Propulsion by Differences of Fluid 19. H.S. Seifert and D. Altman, “A Comparison of Adiabatic and Isothermal Pressure,” Transactions of the Institution of Naval Architects, 30, 1889, Expansion Processes in Rocket Nozzles”, American Rocket Society pp. 390-405. Journal, May-June 1952, p. 159-162. 8. E. Sänger, Raketenflugtechnik, R. Oldenbourg Verlag, Muich-Berlin, 20. C. B. Eom et. al., “High Critical Current Density and Enhanced 1933, in translation as “Rocket Flight Engineering,” NASA TT F-223, Irreversibility Field in Superconducting MgB2 Thin Films”, Nature, 411, 1965. 31 May 2001, p. 558-560. 9. R. Zubrin, “Dipole Drive for Space Propulsion,” JBIS, 70, 2017, pp. 442- 21. A. Matsumodo et. al., “Evaluation of connectivity, flux pinning, and 448. upper critical field contributions to the critical current density of bulk 10. D. Brisbin, “Spacecraft with Interstellar Medium Momentum Exchange pure and SiC-alloyed MgB2,” Applied Physics Letters, 89(132508), 2006. Reactions: The Potential and Limitations of Propellantless Interstellar 22. M. Ranot and W.N. Kang, “MgB2 coated superconducting tapes with Travel,” JBIS, Vol 72, pp 116-124, 2019 high critical current densities fabricated by hybrid physical-chemical 11. K. Tsiolkovsky, “The Exploration of Cosmic Space by Means of Reaction vapor deposition,” Current Applied Physics, 12(2), 2012, pp 353-363. Devices”, The Science Review (Nauchnoye Obozreniye), May 1903, from 23. S. Humphries, Principles of Charged Particle Acceleration, Wiley, New translation “Collected Works of K.E. Tsiolkovsky: Volume II – Reactive York, 1986, p. 453 Flying Machines” NASA TT F-237, 1965, pp 72-117.

Received 5 April 2019 Approved 5 June 2019

152 Vol 72 No.5 May 2019 JBIS JBIS VOLUME 72 2019 PAGES 153–160

DOCKING WITH ROTATING SPACE SYSTEMS

MARK HEMPSELL, Hempsell Astronautics Ltd, 10 Silver Birch Avenue, Stotfold, Herts, SG5 4AR, United Kingdom. email [email protected]

It is a common conclusion that spinning habitats to provide artificial gravity will be an important aspect of humanity’s long term occupation of the orbital space environment. Such systems need docking ports to connect with delivery systems for crews and supplies. System concepts incorporating artificial gravity typically place these ports in a despun section. This means the docking or berthing process is conceptually the same as for non-spinning systems. However in reality the inevitable misalignment between the intended and actual mass properties of the spinning section will create a motion of the docking port that is outside the design misalignment range of the standard docking systems. The spin induced motion could be removed with active mass property control to remove misalignments due to movements within the habitat. Alternatively the motion could be actively compensated for by moving the docking port (or in berthing operations moving the manipulator end effector) so that it remains fixed in space. A simplistic assessment suggests that the trade-off between active control of the mass properties and active control of the motion would seem to favour motion control on a mass and complexity basis. However the design considerations are also effected by the considerations of loads on the despin mechanism which could favour active mass properties control in some circumstances.

Keywords: : Artificial gravity, Active mass properties control, Docking, Berthing

1 INTRODUCTION not significantly improved in the subsequent forty years). The reasoning and extensive referencing behind this conclusion is From Tsiolkovsky onwards it has been appreciated that a spin- provided in Reference 4. ning space system could provide artificial gravity, making the habitable environment more Earth like for its occupants. These conclusions on the impact of Coriolis forces on the rotating space stations featured in both Von Braun’s and the vestibular system were arrived at through terrestrial experi- British Interplanetary Society’s 1950s visions for the future of ments which are clearly complicated by Earth’s gravity acting space flight and also famously in the 1968 film2001: A Space in addition to the spin forces and thus they may not hold true Odyssey [1] which showed spin induced gravity in both space in a zero gravity situation. Experiments conducted in micro- stations and spacecraft which the corresponding novel reveals gravity have shown that the human vestibular system under- were intended to give 1/6 (i.e. lunar) gravity [2]. The creation goes changes in sensitivity after a few days in microgravity and of artificial gravity was a fundamental part of the concept this suggests that higher spin rates may be possible. In 1999, of space colonies as popularised by O’Neill [3] as the whole Hall [5] reviewed the basis of suggested limits, highlighting the premise was the creation of an Earth like environment suitable wide range that existed in the literature but showing between 4 for permanent habitation. and 6 rpm was the consensus view. It may be that much higher rates are possible, for example Hecht et al [6] argued as high However spinning introduces many complications in the as 23 rpm, but even if faster spin rates could be achieved from design so when orbiting stations were first established in the a medical point of view, they are unlikely to be much faster in 1970s they were without spin induced artificial gravity and this practice because of the general practicalities of humans operat- has been the case for every space station since. These micro- ing in an environment with a highly curved motion compared gravity stations have established the detrimental long term ef- with the linear motion of normal gravity. fects of microgravity on the human body, such as muscle wast- age, loss of blood volume and decalcifying of the bones. These This work has considered the range 3 rpm to 0.8 rpm al- lead to the conclusion that for a general population living in though the fundamental conclusions reached are actually in- space on a long term basis, spin induced artificial gravity will sensitive to spin rate. Fig. 1 maps the spin rate to radius for be essential. different gravity levels. These slow spin rates mean that to -ob tain a one Earth gravity equivalent, the radius will have to be Terrestrial experiments in spin induced gravity conducted large compared with current spacecraft. It can be seen that to in both USA and Soviet Union revealed that the Coriolis forc- obtain one gravity requires 224 m at 30 seconds and 894 m at es on the vestibular system (created by the fast spin rates), in- 60 seconds. It may prove possible that humans can avoid the duced nausea in the test subjects. This lead to a design assump- detrimental physiological effects of microgravity with less than tion of one revolution a minute by the Stanford workshop for one gravity although, with no experimental data, this must cur- their space colony design. This was set as a safe “conservative” rently be viewed as conjecture. Even if we make an assumption value given the poor knowledge base at the time (which has that a half earth gravity is sufficient then at 30 second revolu-

JBIS Vol 72 No.5 May 2019 153 MARK HEMPSELL tion the radius is 112 m and at 60 seconds it is 447 m. The inevitable conclusion is that any space system providing artificial gravity to maintain human physiology will have dimensions of hundreds of meters. It follows that the rotating section of such systems will dominate its mass properties.

Any system that is spinning to provide artificial gravity for its inhabitants will need supplies and the means to exchange personnel. So supply spacecraft will need to connect to it; a difficult technology in zero-g that is further complicated if one of the systems is spinning. This issue has been ignored in the past literature on artificial gravity (although there is some un-ref- ereed internet discussion [7]). For example probably the only book dedicated to the subject of spin created artificial gravity was edited by Clement and Bukley [8] and was intended to be a comprehensive major review of mankind’s knowledge on spin induced gravity. Yet this volume has no discussion on the mass properties required for stable spin or the complications of docking with spinning systems. Yet these two issues and their interaction will probably prove the most important practical engineering concern in the realisation of space systems that in- Fig.1 Spin Time against Radius. corporate spinning habitations. for spin induced motion on what is assumed to be a fixed point The particular problem considered here is the discrepan- in space. This raises the question of how much of the misalign- cy between the designed mass properties and the actual mass ment budget can be allocated to the passive spinning system properties. This inevitable discrepancy means the connecting given that the original assumption was that this was not a con- port will be moving relative to the approaching spacecraft with tributing item. With no experience on constructing spinning the spin period of the habitat. And, as will be shown, this mo- habitations, the reality of what can be achieved must remain tion will almost certainly be outside the capability of the dock- speculation. However, for this analysis, we assumed that spin ing standards that have been developed for non-rotating space axis misalignments must be less than 25 mm and angular mis- systems. alignments must be less than 1 degree.

2 STATEMENT OF THE PROBLEM 2.2 Docking Strategy Options

2.1 Docking System Constraints There are three basic strategies by which a spacecraft could -ap proach and make a physical connection with a large spinning There have been several docking systems used over the history object, they are: of human spaceflight. All have been developed for the situation i) The spinning system despins to become 3 axes stabi- where the passive spacecraft is 3 axis stabilised and “drifting” lised for each docking then spins up again to restore so that the misalignments can be reduced to measurement in- artificial gravity. accuracy of the relative positions and orientation, and the pre- ii) The docking spacecraft matches the spin and then cision of the active spacecraft’s reaction control system. This docks along the spin axis at one or the two “poles” means that the docking system’s specified misalignment range iii) The system has a non-rotating “despun” section for is intended to only account for the active side, and the passive conventional docking operations. side is assumed to be the reference against which the misalignments are determined. The first of these options maybe practical for small system and is the method assumed for a design study of a multi-pur- The most recently developed docking standard, the Inter- pose crewed vehicle, called the Scorpion (Fig. 2). The range national Docking System Standard [9], is specified to operate of spin and the corresponding reaction control propellant re- with linear misalignment of 110 mm together with a rotational quired to achieve it are shown in Fig. 3. misalignment of 5 degrees about any axis. The corresponding allowable relative movement are a linear speed of 0.4 m/s and a rotational speed of 0.4 deg./s about the docking axis and 0.15 deg./s about an axis in the orthogonal plane. These values represent the body of experience with docking systems such as the Androgynous Peripheral Attach System used by America on the ISS programme. These values were also used for the alternative proposed standard, the Universal Space Interface Standard [10, 11]. Given that the use of specialist spin docking equipment is unlikely to be practical in the context of a wider orbital infrastructure it is concluded that the existing docking system constraints are unlikely to be relaxed and must be worked within.

Given its size; the spinning space system will be the passive target of the docking operation, which means there is very little scope (if any) in the docking system alignment error budget Fig.2 The Scorpion.

154 Vol 72 No.5 May 2019 JBIS DOCKING WITH ROTATING SPACE SYSTEMS

Fig.3 Scorpion Spin Range and Propellant Required.

This spacecraft is intended to undertake long interplanetary proach is that there is no requirement for a rotating connec- missions and has been configured to spin with a radius arm of tion between pressurised sections. However this approach lim- 50 m and designed to spin up to a maximum of once every 20 its the spinning space system to two connection ports (at the seconds to give a gravity of around 0.5 g, but its main require- poles) it also amplifies the alignment problem as there are now ment is to spin at once every 30 seconds; giving around 0.3 g two mass properties mismatches involved. suitable for keeping its crew in condition for Martian surface exploration but probably not enough gravity to maintain hu- It is therefore probably no surprise that most serious studies man physiology suitable for direct return to Earth. that have considered including spin induced artificial gravity have used the third strategy and incorporated a despun section The Scorpion can vary in mass from around 240 tonnes that includes the docking and berthing ports. This approach when empty and unloaded to around 1000 tonnes fully fuelled has the problem of how to create a pressurised connection be- and loaded and, of course, the amount of propellant required tween the spun and despun sections but the advantage is that, to achieve a given spin rate varies correspondingly. The plots in theory at least, it reduces the problem of docking to the same in Fig 3 show the values for an empty Scorpion and one fully as a non-spinning system; the situation for which the current fuelled and loaded for a Martian mission. They show that a spin docking standards have been developed. However, in practice, up or despin manoeuvre requires in the order of a tonne of the actual system’s mass properties mean these docking stand- propellant, therefore a spin up followed by a despin requires a ards probably will not be able to operate directly in the manner total of around two tonnes. This is acceptable in the context of envisaged by the initial design. an interplanetary mission where it will only be done twice. It would not be so practical if despinning was required on a regu- 2.3 Spinning Systems lar basis, nor would the propellant consumption be practical on the larger systems that would be required to provide close to a The mass properties of an object in Cartesian coordinates are full Earth gravity, with their much higher angular momentum. defined by an inertia matrix: Ixx Ixy Ixz The Scorpion study did look at using reaction wheels to Ixy Iyy Iyz store the angular momentum while despun but the size proved Ixz Iyz Izz unrealistic (far greater than the propellant required) and the gyroscopic rigidity was thought likely to hinder the spacecraft’s Where Ixx, Iyy, Izz are the inertias about the orthogonal ability to perform attitude manoeuvres. The conclusions re- Cartesian axis x, y and z. Ixy, Ixz, Iyz are the cross products that garding the relative mass of the reaction mass would also apply are a measure of the link between the axes, which means that to larger systems, unless the system’s configuration employed a torque applied about any of the axes will cause motion out- two counter spinning habitation sections. side the plane, causing the object to tumble. Thus an object will only have pure spin, i.e. without tumble, about an axis when There are several problems with this approach, but the im- the cross products are zero. All bodies have at least one set of mediate obvious argument against it is the loss of gravity dur- orthogonal axes for which this is true. Stable spin will occur ing docking operations and the associated inconvenience for about the axis in this set which has the highest inertia, and the the inhabitants. Therefore for large habitats it is assumed the larger the ratio between this axis and the other two inertia axes docking will have to be conducted while the system is spinning. (the inertia ratio) is a measure of the stability of the spin.

The second strategy of matching spin axis and rates was fa- The design aim for a system intended to spin is that the in- mously portrayed in the film 2001: A Space Odyssey [1]. Un- tended spin axis has the largest inertia ratio. A further con- fortunately for the purposes of this paper; the film cuts from sideration is to ensure that as far as possible this spin axis is a shot of the Orion Spaceplane matching its spin with the sta- aligned with the design axes, which of course is the set of axes tion while approaching down the rotational axis, to a shot of to which the docking system will also be aligned. the passengers disembarking from the lift at the station’s rim. Thus leaving the manner by which the spaceplane actually con- The author’s experience is with communications satellites nects to the station as an exercise for the audience’s (hopefully that were required to accurately spin at around 60 rpm for the considerable) imagination. The potential advantage of this ap- transfer orbit phase of the mission. Generally it proved possible

JBIS Vol 72 No.5 May 2019 155 MARK HEMPSELL to arrange the satellite’s equipment so the centre of mass was predicted by the mass modelling to be within a centimetre of the design spin axis – that is 1% of the satellite’s dimensions. When the satellites were actually made, these model predic- tions proved quite accurate, although the prediction of the axis angles were less reliable. During the launch preparations the satellite would be placed on a spin table and balance masses added to align the centre of mass and spin axis with the design intention. A one tonne satellite typically could be balanced for around 10 kg of balance masses secured to hard points provided for the purpose.

Extrapolating this experience into the design of spinning habitation systems, the ability to locate the centre of mass through the location of the equipment is probably similar, i.e. within 1% of the system’s dimension. The angular misalignment is probably much less of a problem as, to maximise the inertia ratio, spinning habitation systems tend to lie in the plane orthogonal to the spin axis, which almost reduces the cross products in the design axes to zero. Fig.4 Rockwell Post Apollo Proposal for a Space Base. 2.4 Examples

The drive to maximise the spin inertia ratio can be seen in three examples of concept designs for spinning space bases, the Rockwell post-Apollo programme space base (Fig. 4) the Stanford Torus space colony (Fig, 5) and the Skyfarm concept (Fig. 6). These examples span the likely range of systems which incorporate an artificial gravity section that will continue to

spin during docking operations. The spinning sections of these COURTESY OF NASA/DON DAVIS systems are different in configurations, (dual dumbbell, torus and disk), but their thin sections mean all have an inertia ratio close to two, maximising the stability of the spin.

The first of these examples comes from two competitive Phase B studies for a post-Apollo space station – one by North American Rockwell [12] the other by McDonnell Douglas [13]. These studies were for a post-Skylab twelve-man space station to be launched by a Saturn 5. The design was intended to act as a standard habitation module for crewed spacecraft and space bases. To demonstrate this requirement both studies includ- ed designs for a base with artificial gravity that could house Fig.5 Rockwell Post Apollo Proposal for a Space Base. over fifty people. The habitation section of the North American Rockwell base was composed of four of these modules in a cru- ciform, Reference 10 gives the spin radius as about 35 m so it is unlikely it would provide a full Earth gravity, but it is not clear the degree to which the designers understood the limitations on spin rates. For this paper we have assumed a spin rate of 20 seconds which would give around one third Earth’s gravity.

The total base estimated weight was 420 tonnes, which probably an underestimate given this is almost identical to actual mass of the much smaller International Space Station [14]. While not recorded; the mass of each of the modules would be around 60 to 70 tonnes making the spinning section more than half the total mass of the station.

The second example is the Stanford Torus, which was the most mature and developed concept from the space colony Fig.6 Skyfarm. thinking of the 1970s led by Gerald O’Neil [4]. It was the out- put of a ten week study workshop held at Stanford University. requires a radius of 895 m). Its mass was estimated to be 10 The main habitation section was a tubular torus with a major million tonnes, nearly all of which was radiation shielding. This radius of 830m and minor radius of 65 m. This gives a “floor” shielding was not spun in order to reduce the structural loads. radius around 850 m which would be spun at once a minute The mass of the spinning section is around 200,000 tonnes. The to provide one Earth gravity (there is a slight inconsistency in docking provisions were to be provided by two (north, south) the report [4] which acknowledges that to achieve 1 g at 1 rpm despun sections connected with a 15 m bearing.

156 Vol 72 No.5 May 2019 JBIS DOCKING WITH ROTATING SPACE SYSTEMS

The third example is the Skyfarm, as outlined in Reference Table 2 Estimated Centre of Mass Misalignments 15. This was not intended as a human habitation system; as the Moveable mass Movement C of M offset name suggests, it is an agricultural facility exporting food to (tonnes) (m) (m) the wider space economy. It has a spin radius of 3.5 km and with one revolution every 108 seconds provides a range of Post Apollo Base 5 35 0.42 gravity levels from 1.03 g to 0.36 g on 36 working floors. It has Stanford Torus 100 830 0.40 a mass of 305 million tonnes; almost all in the spinning section. Skyfarm 20,000 1,000 0.07

A summary of the key features of these examples is given in Table 1. ules at 4 tonnes each = 2,800 tonnes), the crew of 10,000 people (1,000 tonnes) and to this must be added the movements of TABLE 1 Key Parameters of the Examples water, nutrients and the subsequent growth of plants. The total mass that moves on a daily basis is in the order 20,000 tonnes Mass Radius Period Gravity but the movement will be random and over many different ra- (Ktonnes) (m) (sec) (g) dii. Thus creating a design case is more difficult in such a spec- Post-Apollo Base 0.42 35 20 0.33 ulative concept. However this study assumed a displacement of Stanford Torus 10,000 830 60 ~1 1 km which would cause an overall centre of mass movement of 7 cm. Skyfarm 305,000 3,500 108 1.03– 0.36 All these design cases are extreme in order to avoid any operational constraints. They are summarised in Table 2. 2.5 Misalignments The movement of mass around the spinning habitat will also Just as with smaller satellites, if any of these systems are ever alter the inertial matrix such that the actual spin axis has a mis- realised, then the reality of design constraints and build er- alignment to the design spin axis causing a periodic motion in rors will mean the actual mass properties will not match the both the position and the angle of any docking ports. required design axes. It is possible that the higher percentage of structure mass with its inherent symmetry, and the larger This potential angular misalignment was modelled by tak- number of components with greater repetition for redundancy ing the assumed moveable mass along the spin axis (and thus would mean the contribution of the design mismatch could be rotating the spin plane) to the maximum possible in the habi- smaller as a percentage. However the build errors are likely to tat. An extreme case, but again a design case that means there be higher because, if construction costs are to be kept to an would be no constrains on the uses of the habitat. The resulting economically realistic level, they will be built more along the distance misalignment is a consequence of both the misalign- lines of civil engineering than precise aerospace engineering. ment angle and the distance of the docking port from the cen- This imbalance will need to be corrected by balance masses and tre of mass. The results of this modelling are shown in Table 3. in the absence of practical experience on such large systems it is suggested a 1% allowance should be incorporated in mass Even though the modelling was unrealistically extreme the budgets. resulting angular and lateral misalignments are very small, less than a tenth of a degree and, except for the post-Apollo Base, Even with static balance masses, it is unlikely that the axes sub-millimetre in displacement. In the case of the post-Apollo can be aligned to below 5 cm; in the case of the post-Apollo Base the motion is almost 5 cm which is significant compared station that is 0.15% of the spin radius, for the Skyfarm it would to the allowed docking misalignment, but is an order of magni- be 0.0015%. However even if the static balance masses could tude lower than the displacement due to the movement of the reduce the centre of mass mismatch to a negligible factor there centre of mass. is still the matter of mass changes due to operations. Inherent in its function as a habitation area are the changes in the centre It is concluded that spin angle movement is not likely to of mass due to movement of the crew, supplies, equipment and prove a problem that requires any special provisions. This is other movables within it while it is operating and these will also because all the chosen examples were configured to maximise alter the mass properties in a dynamic way. the inertia ratio which inevitably leads to a small motion range along the spin axis. To get an estimate of the likely impact of this dynamic component of the mass properties three design cases were gener- It is concluded that the problem with docking is confined ated. For the post-Apollo station take the case of all the crew to the motion due to the difference between the real centre of (50 – 60 people) assembling in one module – say for a crew mass and the design centre of mass used by the location of the meeting. That would place around 5 tonnes at 35 m that would docking port. The three cases suggest that as systems get larger, move the centre of mass of the whole station 0.42 m. For the the movement reduces because of the ratio of moveable and Stanford Torus take the case of every single inhabitant moving to a single point on the rim – for say a Presidential inaugu- ration. That would be around 100 tonnes at 830 m and again TABLE 3 Angular Misalignment that would move the centre of mass by about 0.4m. This may Mass Offset Angle Port Motion be larger than a real system would experience as the Stanford (m) (degrees) Distance (m) Radius Torus mass estimate may be be a significant underestimate. (mm) Post-Apollo Base 4.5 0.09 30 45 For the Skyfarm; the system produces 7,000 tonnes of food Stanford Torus 60 4x10-5 120 0.1 a day, which moves around the facility. There are also mass -4 movements from the transport of logistics modules (700 mod- Skyfarm 100 1x10 500 0.9

JBIS Vol 72 No.5 May 2019 157 MARK HEMPSELL total mass reduces, and the moveable mass motion becomes Using the assumptions that the active balance mass is locat- more randomised. However in every case this motion exceeds ed at the habitat radius and t = 10 minutes (600 seconds). Then the standard docking port requirements by an order of mag- for the three example cases the mass, the mass flow rate and the nitude. velocity are given in Table 4.

3 DOCKING CONTROL OPTIONS Table 4 Characteristics of Active Mass Property Control System 3.1 Active Centre of Mass Control Mass Mass flow Velocity (tonnes) (tonnes/s) (m/s) One way to resolve this movement is to actively control the Post-Apollo Base 6 0.008 0.2 mass properties through moveable balance masses, either solid masses on rails or fluids moved between tanks, the latter a prin- Stanford Torus 120 0.167 4.5 ciple used by some aircraft which move fuel between tanks to Skyfarm 24,000 33.33 9.16 adjust trim. The aim being to counter the movements of mass within the habitat by actively adjusting the balance mass to maintain the design mass properties. While these results are based on many speculative assumptions, they are thought to provide a realistic indication of the This active mass property control system could also incor- magnitude of any real mass properties control system. In the porate the masses required for static balance. This would make context of the habitats they control, they are a significant, but sense; as the static balance masses will need to be added after probably easily realisable, engineering solution. the spinning section is built require the same sensors as needed are by active control to determine the location of those masses. 3.2 Active Motion Control So one mass balance system could combine both static and active mass properties control. However the static balance mass While active mass properties control would seem to offer a vi- provides the centre point for the subsequent movement of the able approach to the spin axis misalignment problem, in ab- non-fixed masses, thus the balance mass required for the active solute terms the systems involved would be large and heavy. control has to be on top of any static balance mass requirement. The alternative is to allow the movements of the spin axis and compensate for that movement relative to the approaching If the control masses radius (rb) are located at the same radi- spacecraft by active motion control of the docking or berthing us as the habitat radius (rh) then it follows they would need to system. be the same mass as the moveable mass (mm). A shorter radius increases the mass and larger reduces it; thus the active balance A docking system has two components; a capture system mass (mb) required will be found from: that takes out the misalignments to make the initial connection, it then draws the two systems together so their mounting (1) faces meet and the second hard dock latching system can en- gage to give the connection its structural strength and if required make the air tight seal. So it is the lack of alignment This gives a value just for the balance mass itself. It would range in the capture system that creates the problem for the need systems that enable it to be moved. For example if a fluid spinning habitat case. mass was used it would require tanks, connecting pipes and pumps. For this study a 20% allowance on top of the balance It is not practical to increase the range of the capture system mass was assumed to cover this handling equipment. with the existing standards. The capture systems are located inside the mounting ring, which is a little over a metre in di- Locating balance masses on structures beyond the habitat’s ameter. When factors such as the need for an internal transfer radius has the attractions of reducing the mass required while corridor are considered there is not enough space for mecha- at the same time increasing the inertia ratio and thus improving nisms that can increase the misalignment range by a factor of spin stability. So this is an attractive design solution for static five or more. It is concluded that the only route to making the balance. However for active balance control there is a downside existing docking standards work is to place them on a six de- as the further away the balance mass is, the faster it will need gree of freedom mechanism, such as a Stewart Platform, which to move to track the habitat’s mass movements. The minimum counteracts the spin motion; holding the docking port steady distance of movement would be along the radius but this would in space for the active system to connect as shown in Fig. 7. only give control along one axis. So to get two axis control the movement is more likely to be around the circumference with The approach will require a means to maintain a pressur- a distance of travel a quarter that circumference; a difference ised environment once the connection has been made, either from a pure radial movement of π/2. The minimum time for by a second clamped seal (shown in Fig. 7) or a flexible tunnel the moveable mass to reach the extreme cases (t) is likely to between the base and the moving docking ring. Otherwise this be in the order of minutes which is greater than the spin rate approach would seem to be straightforward to implement as (reducing the complexity of the control system). From this the the motion needs only to match a predictable cycling spin mo- rate of balance mass movement is given by: tion. The mass would be considerably less than an active mass properties control system and, given there would be more than (2) one docking port, this approach would give a redundancy to the overall system that active mass properties control would And the velocity (v) is given by: not.

(3) The alternative to docking is berthing; a technique that has been proven on the International Space Station (ISS); first

158 Vol 72 No.5 May 2019 JBIS DOCKING WITH ROTATING SPACE SYSTEMS

Fig.7 A Docking Port Mounted on a Stewart Platform (retracted and extended). with the Japanese HTV and latterly with the Dragon and Cyg- 4 Spin Despin Connection nus supply vehicles. A berthing operation involves the active spacecraft holding station relative to the passive target while There is a second consequence of any mass properties mis- a manipulator physically connects them and then draws them alignment in the spinning section of a space habitat which is together to hard dock. So the manipulator serves the same the bearing loads it would create on the despin mechanism. It function as the capture system in a docking operation, it takes this rotating mechanism that form the connection between the out the misalignments and brings the two objects’ mounting spinning and non-spinning sections. faces together to make the structural connection. In theory the manipulator could be on either the passive or active side, but all The despin mechanism would be a critical element in any berthing operations on the ISS have used the station’s Remote habitat which incorporates a despun section for its spaceport Manipulator System. and it represents a difficult engineering challenge. It needs to power and maintain the continuous controlled rotating con- This is a potentially attractive approach. Given the facility nection between two sections, each of which will have a mass will have manipulators for berthing and other operations any- from hundreds to hundreds of millions of tonnes, for decades, way, it has no mass impact. Indeed given the attachment port which in and of itself is a non-trivial exercise. However it also has only a berthing function it would even be lighter than the must incorporate safe failure modes, be maintainable and allow options requiring a full docking capability. However it is not a pressurised transfers between the two sections. The latter func- direct application of the operational approach proven on the tion requiring either a rotatable seal or a transfer compartment ISS. ISS berthing is controlled by astronauts, the manipulator that alternately connects with the two sections spinning up and having no part in the guidance of its end effector to the grapple down in between. These engineering challenges (which are all point. It is unlikely that with the spin induced motion added derived from the inherent functionality), are significantly com- this would remain a task that could reliably be left to human plicated by mass properties misalignment. control which means the attachment process would need to be automated. This may require some additional sensors (although Any offset of the rotating objects’ centre of mass creates a existing cameras and target recognition may be enough) the force in the bearing constraining it. Even assuming the static main technical uncertainty is with the control software. This balance masses do a perfect job; the moments created by the will require advanced image recognition and control software moveable masses within the habitat are considerable and giv- amounting to Artificial Intelligence and would represent a sig- en the other requirements on the despin connection it may be nificant advance in berthing technology.

There is a second impact on the wider infrastructure. To enable berthing the approaching spacecraft must have a grapple point for the manipulator’s end effector. The first implication of this is that all transport systems that interact with the spinning system will need to have grapple points fitted, regardless of whether berthing was intended as the primary means of system to system connection. With the ISS, support systems have always been equipped to do either berthing or docking, never both. A wider infrastructure that has spinning systems within it is likely to require all spacecraft to have the provisions for berthing even if they are primarily intended to employ docking (Fig. 8). The second implication is that the grapple points that are fitted will need to be an infrastructure wide standard following the same logic as with standardised docking/berthing ports [11]. Fig.8 Grapple Point (Indicated) Located for Berthing Operation.

JBIS Vol 72 No.5 May 2019 159 MARK HEMPSELL that a better system level solution is to incorporate active mass However if consideration of the loading on the despin properties control to reduce the bearing loads. In this event the mechanism proves to be an issue, then that could raise the de- problem of spin induced movement of docking ports is also sirability of using active mass property control to reduce the resolved almost as serendipity. loads on the bearings. This system level consideration could move the optimum strategy to active mass properties control. 5 Conclusions The current uncertainties about the two fundamentals of ar- The subject of handling the discrepancy of a spinning system’s tificial gravity, the level of gravity required and practical spin centre of mass in the context of creating artificial gravity for its rates for human occupation, mean the actual design require- habitation does not seem to have received serious examination. ments are not understood. It follows that no definitive conclusions as to the best strategy can be reached. Further the actual Discrepancies between the desired mass properties that are approach to resolving the problem will probably differ from due to design constraints and build inaccuracies would need system to system. As systems get larger the scope of the motion to be addressed by static balance masses, which would have to reduces, but the loads on the bearings increase and the mass be accounted for in the design process. But operational activity and complexity of any mass properties control system also in- within the habitat can produce mass properties misalignment creases, all leading to a complex system level trade-off to find that would induce motion in the docking ports that takes them the optimum solution. considerably outside the misalignment range of the standard docking systems that would be fitted to incoming spacecraft. Although not examined in past literature, mass properties misalignment of spinning habitats is a significant engineering Two strategies have been identified to resolve this problem, issue. It has a system level impact and needs to be addressed at either actively control the spinning section’s mass properties or the concept feasibility level in the system development life cy- actively control the motion of the port. An initial assessment cle. While it is not possible to identify a best strategy applicable suggest all of these options are practical and realisable, but to all systems, all the options examined would seem to be viable the motion control would appear to offer a much lighter and and realisable. Therefore, there is no concern that this issue ef- cheaper approach. fects the viability of spin-induced artificial gravity.

REFERENCES 1. 2001: A Space Odyssey – film, director S Kubrick, MGM 1968 https://worldbuilding.stackexchange.com/questions/134530/what-is- 2. A.C. Clarke, 2001: A Space Odyssey, Hutchinson, 1968 the-best-design-for-docking-onto-a-rotating-space-station 3. G. K. O’Neil, The High Frontier: Human Colonies in Space, Jonathan 8. G. Clement and A. Bukley, Artificial Gravity, Springer Science + Cape, 1977 Business Media LLC 2007 4. Ed. R. D. Johnson and C. Holbrow, Space Settlement: A design study, 9. International Docking System Standard (IDSS) Interface Definition University of the Pacific 2004 (reprinted from 1977) Document (IDD) Revision C, 20 November 2013 5. T. Hall, “Artificial Gravity and the Architecture of Orbital Habitats”. JBIS 10. USIS Technical Requirement Specification Draft F (this can be Vol 52, pp 290-300, 1999 downloaded from www.hempsellastro.com or www.usisassociation.org) 6. H. Hecht, E. L. Brown, E. L., and L. R. Young. “Adapting to artificial 11. M. Hempsell, “Creating a Universal Space Interface Standard”, JBIS, Vol gravity (AG) at high rotational speeds,” Proceedings of Life in Space for 69, pp 163-174, 2016 Life on Earth, 8th European Symposium on Life Sciences Research in 12. D. Baker, “Space Station Situation Report -1 The North American Space. 23rd Annual International Gravitational Physiology Meeting, 2-7 Rockwell Proposal”, SpaceFlight Vol 13. 8 pp 318-334, Sept 1971 June 2002 13. D. Baker, “Space Station Situation Report -2 The McDonnell Douglas 7. Examples (all accessed 15 April 2019): Proposal,” SpaceFlight, Vol 13. 8 pp 344 351, Sept 1971 https://space.stackexchange.com/questions/20234/docking-options-for- 14. https://en.wikipedia.org/wiki/International_Space_Station (Accessed a-rotating-space-station 4th Feb 2019) https://www.quora.com/How-would-you-dock-with-a-centripetal- 15. M. Hempsell, “Skyfarm: Feeding a Large Space Population”, JBIS, Vol 70, space-station pp 3-11, 2017

Received 20 April 2019 Approved 27 April 2019

160 Vol 72 No.5 May 2019 JBIS JBIS VOLUME 72 2019 PAGES 161-171

MARTIAN DUST: the Formation, Composition, Toxicology, Astronaut Exposure Health Risks and Measures to Mitigate Exposure

JOHN R. CAIN Independent Consultant, Hookstone Chase, Harrogate, N Yorkshire, UK email [email protected]

This paper reviews the geology of Mars and the physical properties of the Martian dust including its distribution, particle size and chemical composition. The toxicity of the dust including its characterization is discussed with the associated exposure health effects. Mars has a core, a mantle and a crust of mainly ferrous sulphate, magnesium/iron and silicon dioxide respectively. Various models have been described to account for the Martian regolith/dust formation, for example, by aeolian processes over time. The models are being modified as more data is collected. The sources of the dust and its airborne transportation are due to ordinary atmospheric settling, from dust storms and dust devils etc. The dust settles and adheres on surfaces by one or more of a combination of forces. Both dust transport and adhesion can increase the risks of astronaut exposure in particular when the spacesuit is removed in the airlock and high airborne dust levels may be reached. If the dust particles are inhaled, ingested, absorbed through the skin or have contact with the eyes then significant health effects could occur. Martian dust is known to consist of a number of toxic substances including respirable crystalline silica, hexavalent chromium, arsenic etc. and perchlorates. Inhalation of the smaller particles, below 10 µm down to 0.1 µm, into the deeper reaches of the lung may cause oedema, fibrosis and cancer. The setting of an appropriate permissible exposure limit for the Martian dust of 0.05 mg/m³ is proposed. The mechanical astronautical hygiene techniques that are being developed to mitigate exposure to the Martian dust are outlined, for example, the use of an electrostatic precipitator to collect contaminated dust from the air and local exhaust ventilation to contain and exhaust contaminated air. The need for individually designed spacesuits that are flexible and ergonomically designed is discussed.

Keywords: Martian dust, Dust characteristics, Toxic substances, Exposure respiratory disease, Astronautical hygiene, Mitigation

1 INTRODUCTION billions of dollars into developing a means to reach the planet, explore the surface and to eventually establish settlements. Humankind has always had a fascination with Mars, the Red However, such ambition will not succeed unless the astronaut Planet. In prehistoric times it was one of the five planets visible exposure health risks are identified and sustainable measures to the naked eye. The ancient Greeks and Romans saw Mars taken to mitigate such exposure [1]. Exposure to Martian dust as the god of war. During the seventeenth century, Mars was will be one of the major health risks that will need to be inves- first observed with a telescope and drawings described dark tigated and understood prior to the journey to Mars and the areas reminiscent of vegetation. With more powerful tele- subsequent establishment of settlements [2]. There is therefore scopes, lines were seen on the planet’s surface and identified as the need, as discussed in this paper, to examine how the char- canals (or canali as called by Schiaparelli). Were these canals acteristics of the dust including the formation, the composition evidence of life? The stage was now set for the science fiction and the physical properties may affect the health of the astro- writers such as H G. Wells and Edgar Rice Burroughs to popu- nauts and others that will live and work on the planet [3] [4]. late the planet with weird and terrifying creatures. 2 MARTIAN GEOLOGY In 1965, Mariner 4 sent back the first close-up images of Mars; they revealed a world of craters as on the Moon. The twin 2.1 The Martian Core Viking Mars Landers transmitted the first images from the actual surface of the planet and it showed a rocky environment Planets have a core, a mantle and a crust each of which varies with a surface covered in red dust. Exploratory rovers to Mars in composition. These three layers are differentiated by their then followed including Pathfinder, Sojourner, Spirit and the densities or temperatures and by extrapolation from laboratory more recent Curiosity that touched down on the surface in Au- experimentation on simulated computer modelling. Mars, like gust 2012. Curiosity continues to send photographs from the other planets in the Solar System consists of a metallic core; surface and all reveal a dusty, rocky terrain. future exploratory trips to Mars to study the seismology of the plant will provide more accurate data on the core structure and There is an impetus for humans to explore Mars for national, composition [5]. financial and economic reasons and to satisfy human curiosity. The race to be the first to land on Mars is being spurred From evidence collected from the “Pathfinder” lander and on by wealthy individuals such as Elon Musk who has invested other landers from 1997 onwards it has been seen that the core

JBIS Vol 72 No.5 May 2019 161 JOHN CAIN of Mars consists of a “eutectic” mixture of ferrous compounds sistent average chemical and mineralogical composition over i.e. iron (Fe) and ferrous sulphate (FeSO4) [6]. A eutectic sys- the whole of the planet i.e. mainly silica and ferrous oxides. The tem is a homogenous solid mix of atomic or chemical species chemical composition of the dust varies in the concentrations of such as iron that form a super-lattice from liquid. The forma- the chemicals present and it is based on the type of topography, tion of such a system influences the structure and composition on the weathering processes and on the influence of past vol- of the Martian mantle and crust. The red ferrous compounds canism and meteorite impaction [16] [17]. Table 1 outlines the account for 15% to 50% of the planet’s surface and are respon- overall properties of the dust based on the findings of Perko et sible for the red coloured dust. al. [15]. The chemical composition of the dust is mainly silicon dioxide (43.4–48.6%) and ferrous oxide (17.5–18.2%). The sus- 2.2 The Martian Mantle and Crust pended dust in the atmosphere is mainly below 10 µm with particle size distributions on the surface containing little between Evidence from meteorites show a mantle that has magnesium 20 µm to 60 µm. There is therefore the high risk of health effects and iron at a 75% molar ratio whereas for Earth it is 89% [7]. from the inhalation of the smaller dust particles. Data from the “Pathfinder” lander indicates that the Martian mantle may overlay a core 1300-1450 km in radius and a litho- Dust particle shapes vary from disc shaped (impact ratio of sphere thought to be over 250 km thick [8]. 1/10) to tabular, irregular and round based on models of sus- pected dust settling [18]. Because of the aerodynamic proper- The crust of Mars covering the planet’s surface contains ties of the dust, the shapes of the dust particles that settle on the regolith and the dust with rich deposits of silicon dioxide surfaces will influence the concentrations of the dust deposited (SiO2) (andesites, weathered basalts) and iron. The Martian around settlement modules, on spacecraft, on Rover exposed regolith is a layer of loose material that covers the rocky surface surfaces and on spacesuits during extravehicular activity (EVA) and contains the dust remnants of broken rock and other com- i.e. activity done by an astronaut outside a spacecraft [19]. The ponents. The Martian dust is the finer fraction of the regolith accumulation of dust settling on electrical circuits, cameras and it is the dust of particle sizes between 1.0 µm to 10 µm that and other equipment could cause a failure in one or more sys- are of major concern to the astronauts because of the potential tems resulting in the potential for risks to astronaut health [20]. inhalation exposure health effects. On Earth, the inhalation of silicon dioxide dust may result in silicosis, a life-threatening Furthermore, the formation, shape size and composition of lung disease [9]. the dust in dust devils and dust storms in and around the settlements will relate to the deposits of dust on surfaces that could Andesite found on Mars is an extrusive, igneous volcanic be distributed by the astronauts when they carry out EVA and rock with an aphanitic to porphyritic texture; it is intermediate disturb the regolith. This dust therefore will need to be con- between basalt and granite with ranges from 57% to 63% SiO2 trolled to low levels [21]. Liu et al. have indicated that cyano- [10]. Igneous and aphanitic rocks typically form from lava and bacterial crusts could be used for controlling the fine dust and porphyritic rock. Developments in the conditions of the rocks to resist sand wind erosion in particular in enclosed spaces and during cooling of the magma will change very quickly to form in habitats around each settlement [22]. various textures including phaneritic, aphanitic, porphyritic, glassy, pyroclastic and pegmatitic [11]. The dust has the ability to adhere to surfaces due to its chemical composition for example, by chemical bonding, cementa- Because of the structure of the core and mantle of Mars, the tion and by van der Vaals and electrostatic forces. To reduce crust composition is mainly similar over the whole planet even significant accumulation of dust in and around a settlement though there will be localised differences due to weathering and during EVA, its removal will therefore need to be a regular [12]). Evidence from Maltagliati et al. [13] has shown that wa- task using various methods such as vibration techniques, direct ter could be an important factor in shaping the Martian surface brushing etc. [23]. and that the sub-surface regolith could be a substantial reser- voir for Martian water [14]. 3 MODELS FOR THE FORMATION OF MARTIAN REGOLITH/ DUST 2.3 Physical properties of Martian dust 3.1 Introduction Perko, Nelson and Green [15] have reviewed the papers on Mar- tian dust composition. They note that Martian dust has a con- Meteorites on Earth have so far been the main source of anal-

TABLE 1 The average Martian dust diameters, particle sizes, chemical and mineralogic compositions based on the findings of Perko et al. [15] Dust Parameters Dust sizes and Composition 1. Average diameter Suspended dust in atmosphere: 0.06 µm, 0.8 µm, 2.0 µm, 0.8 µm–5µm. Surface drift material: 0.14 µm–>2.0 µm. 2. Particle size distribution on surface Individual particles up to 2mm diameter, 1–4 mm diameter for agglomerates. Particles from 20 µm–60 µm lacking on Mars, 2µ–10 µm measured around the Viking from the surface area of samples. Up to 30 µm–40 µm diameter on Pathfinder lander wheel abrasions. Sand sizes >100µm; the basalt particles are quickly changed to silt and clay-sized particles by Martian aerobic processes.

3. Chemical composition 43.4–48.6% SiO2 / 7.2–8.3% Al2O3 / 17.5–18.2% Fe2O3 / 6.0–7.5% MgO / 5.8–6.3% CaO / 0.6–1.1% TiO2 / 0.1–0.3% K2O 1.3–2.2% Na2O / 5.4–7.2% SO3 / 0.6–0.8% Cl / 0.1–WT/% H2O / 50–700 ppm CO2 4. Mineralogic Secondary weathering products of mafic igneous rocks (i.e. Martian basalts) perhaps from palagonization. Best consensus model is iron-bearing smectite clay stained by a magnetic ferric oxide mineral (possibly nanophase maghemite). Calcium and magnesium sulphates and chloride salts also present.

162 Vol 72 No.5 May 2019 JBIS MARTIAN DUST: the Formation, Composition, Toxicology, Astronaut Exposure Health Risks and Measures to Mitigate Exposure ysis in studying the early origins of Martian dust [24]. The me- will be important when undertaking an exposure risk assess- teorites were blasted off the Martian surface due to bombardment for determining the exposure risks to health. ment impacts millions of years ago and landed on Earth. An alternative model to complement the above processes, Meteorites are categorised and/or classified according to indicates that both the physical (e.g. wind, dust devils), chemi- their composition, the degree of any alterations, their likely cal (redox reactions of Fe2+/Fe3+) and the re-deposition of dust source and by the constituent metallic sub-groups for example particles (~2 µm in size) are responsible for the formation of aluminium. Stones account for 94% of meteoritic falls compris- rock coatings that cannot be distributed by wind [29]. ing chondrites (calcium/aluminium) at 86% and achondrites at 8%. The Martian meteorites found on Earth have been well In addition, Foley et al. [30] indicate that Mars volcanic studied by scientists because they have been associated with steam vents and hydrothermal activity are responsible for the the potential for life on Mars e.g. Meteorite ALH84001 is classi- production of crystalline Fe and S-bearing minerals such as he- fied to be the oldest meteorite to have come from Mars with ev- matite that may form in high concentrations and settle as dust. idence that could indicate that there was life on the planet [25]. 3.3 Weathering, Chemical Reactions and Cementation As well as the study of Martian meteorites, direct photo- graphic and spectroscopic evidence of the Martian surface Dust movement due to weathering may result in chemical in- indicates large shield volcanoes high in iron oxide (FeO) mag- teractions between the dust particles to form a cemented soil mas and a higher content of volatile material than on Earth for that is not affected by aeolian processes. Over time, hardened example, sodium, potassium, phosphorous and rubidium [26]. rock may form over a chemically leached layer of rock. It has These magmas will have contributed to the early formation of been suggested by McSween et al. [31] [32] and based on sim- Martian dust. ulated laboratory studies that coatings of drift material may be responsible for rock variations. The effects of the dust coatings It has been possible to study several models for the recent dust on the rock chemistry indicate that rock dust wedges would formation on Mars. This has involved analysing the monitoring have a silica content of up to 56% [33]. Dust from the wedges data obtained from the “Viking” and “Pathfinder” landers and therefore, could cause significant exposure health effects if it from other Martian surface probes together with the testing of was inhaled over a long time period for example, during the samples to study the possible chemical composition of the dust maintenance of spacesuits following EVA. [27]. An analysis of the spectroscopic properties of chemical mixtures has also been studied [28]. The overall data has shown There are chemical reactions within the dust aggregates that that the recent formation of the Martian dust is mainly due to turn the soil into a cemented soil bonded together by sulphates, weathering processes, together with the effects of chemical re- ferric oxides or oxyhydroxides. These chemical bonds could be actions and cementation within the regolith/dust rather than by broken down but would require more than the action of dust meteorite bombardment which now plays a very small role. devils or wind in particular if the dust particles had been un- disturbed for a long time period. Chemical binders are reactive 3.2 Dust Formation Due to Weathering Processes within the dust and produce a strong cement binding between the ferric, sulphate and silica components of the dust. Such a Bishop et al. [29] have proposed a model for the recent forma- binding effect would reduce the potential quantity of dust re- tion of the Martian regolith and dust based on the mixing and tained on an astronaut’s spacesuit. distribution of the dust components over the planet’s surface and the reactions between the various chemical components of Murchie et al. [34] have shown that the composition of the the dust. This model proposes that the dust is not formed direct- cemented deposits can be identified based on the structure of ly from local rocks but as a result of global and remote weather- the rocky outcrops and with the crusted regolith possibly bro- ing processes. The dust particles distributed via aeolian process- ken. This could be for example, a cemented coat formed of 10 3+ es will fall on the rocks to form a physical dust coating and also – 15 wt/% Fe2O3 in nanophase, Fe oxyhydroxide in a silicate drift deposits between rock surfaces. The drift dust deposits and – sulphate matrix. This would be similar to that of rocks photo- rock coatings will contain larger aggregates of dust particles that graphed by “Curiosity” and are coated in red dust. will be held together by electrostatic or physical forces. 3.4 Water and Dust Movement The wearing of low dust retention spacesuits should ensure that minimal dust is carried to the airlocks following EVA. Liquid water on the Martian surface will have some effect on However, dust devils and heavy winds with 1 µm–3 µm sized the dust especially if it was free flowing and in sufficient quan- dust particles could lift the surface material and distribute it tity to prevent the cementing of the dust aggregates [35]. The locally to settle on and around settlement modules and surface aggregate particles contain a mixture of for example, iron ox- equipment. If these smaller particles become airborne within ide, sulphates, oxyhydroxides and silica grains as well as clay. the settlement and were not effectively removed by extraction, These aggregate particles in the dust will vary in size from 1.0 then these particles could be inhaled into the deep-seated re- µm to10.0 µm [36]. The smaller 1.0 µm particles and less will be gions of the lung with the potential to cause significant health of nanoparticle size and highly toxic if inhaled or ingested [37]. effects over time [3]. 4 DUST FACTORS RELATING TO EXPOSURE HEALTH RISKS If this dust formation model is proved to be the case, then it is highly likely that the dust that collects on an astronaut’s 4.1 Wind Forces and Dust Storms spacesuit during localised EVA will have been formed from a different region of the planet and therefore will not have similar Head has stated that the major dynamic forces shaping the Mar- physical and chemical properties to the locally generated dust. tian surfaces, the crust and lithosphere of Mars over time (there Information on the characteristics of the local dust therefore have been three major time periods on Mars i.e. the Noachian

JBIS Vol 72 No.5 May 2019 163 JOHN CAIN

(~4.65 to ~3.7 Gyr), Hesperian (~3.7 to ~3.0 Gyr) and Amazo- the appearance of sand dunes and ripples e.g. as in Meridiani nian (~3.0 Gyr to Present) are represented by specific geological Planum. During such storms the airborne dust will settle on processes [38]. These processes are linked to interaction with astronaut spacesuits unless they are sheltered inside a Rover the atmosphere e.g. aeolian, polar and they are also linked to or module. interaction with the hydrosphere (e.g. fluvial, lacustrine) with the cryosphere (e.g. glacial and periglacial) or with the crust Wind processes dominate the environment of Mars. Despite lithosphere and interior (tectonism and vulcanism). the carbon dioxide atmosphere, the winds are capable of creating heavy dust storms (average surface pressure 6.5 mbar). Christensen et al. [39] [40] have indicated that the Martian Because of the changing nature of the windspeeds and the highlands are mainly basaltic in composition. This may have direction of the winds including the formation of dust devils, resulted from volcanic and plutonic origins. Tectonic process- these will present the greatest danger to those astronauts out- es on the Martian surface can produce wrinkles, ridges, lobate side the modules and to equipment when it settles in particular scarps and fold belts together with large windblown ripples or the solar panels responsible for providing energy for a settle- sand dunes that may occur near them. Local topography may ment and supporting equipment. The deposition of dunes and influence the wind and resultant ripple patterns [41]. The pres- aeolian features such as dune-forms and wind-streaks could ence of liquid water in the past has resulted in fluvial landforms entrap astronauts who unknowingly ventured into those areas with valleys with fretted channels and alluvial valleys e.g. as in where they occurred. Work in such areas for example, during Elysium with outflow channels [42]. The landforms produced the digging of trenches, during the taking of dust core sam- by flowing water in the past could have caused a build up of un- ples and during the transport of dust from mining procedures disturbed dust and soil which could settle on astronauts’ space- would need to be assessed to determine the most effective and suits if it was disturbed and subsequently adhere resulting in a efficient measures to control the potential dust exposures [49]. high potential for equipment damage, reduced mobility and a risk of exposure when the spacesuit was removed. 4.2 Carbon Dioxide Settling

The topography, landforms and associated natural processes In those areas of the planet (i.e. Tharsis, Arabia and Elysium) in and around the Martian settlement modules will be firmly that are situated in the middle and low altitudes, the cold night established but they will need to be continuously assessed to temperatures may cause atmospheric carbon dioxide to settle determine any changes to the type and composition of the dust as surface frost. The temperatures will rise during the day and that could increase the exposure health risks. For example, the become colder at night. As the frost evaporates in the morning, soil and dust on lake plains and near glacial and periglacial ice the dust will not combine with other grains of dust to make may have carried soil some distance and deposited it in specific larger particle sizes. In these areas, this failure to form larger areas that could accumulate as deposits and entrap vehicles and particles will lead to erosion of the surface [50]. The prevention the astronauts working in those areas [43]. of the combining of dust grains over time will ensure that the dust particles remain small and therefore subsequently pose a Martian dust has a constant average composition over the risk to health if inhaled inside the airlock [51]. But there is also whole planet e.g. consisting of iron rich clays, palagonite and the potential that any frost that settles on a Rover and on min- haematite. Atmospheric settling of the dust is caused as a result ing equipment will cause dust deposits to settle out, be retained of gravity, vertical eddy mixing and ice crystal nucleation. The and damage exposed equipment. The kinetics of the carbon di- settling of the dust could be disturbed by dust storms. Up to 100 oxide ice formation and the melting of the ice as related to the dust storms can occur in a number of areas with the most oc- temperature, dissolution rates and activity energy will require curring during the Southern Summer Solstice at perihelion [44]. further studies to assess whether the potential acute and long- The amount of dust loading during a Martian dust devil storm term exposure health risks are significant or not [52]. could exceed 700 times that of ambient background dust levels. 4.3 Acid Fogs Dust storms of varying strengths are a frequent occurrence on the surface of Mars and they distribute billions of tons of Acid fogs occur on Mars from a build-up of airborne acidic dust particles into the atmosphere. They thereby contribute to vapours arising from volcanos and in this respect, they are sim- the surface distribution and accumulation of the dust when it ilar in formation to those that are found on Earth. These Mar- settles. During EVA, vehicles, exposed astronauts and equip- tian fogs, containing weakly acidic vapours and liquid when ment will become contaminated with the dust particles during they condense on surfaces, will contribute to the weathering and following the storms. It will be essential that the dust that of the surrounding rocks over time [53]. Exposure to the acid settles on the spacesuits doesn’t clog equipment being handled vapours in the weak sulphuric acid concentrations as found on or settle on the spacesuit glove/arm interphases in sufficient Mars is unlikely to cause harm to the astronauts if there is skin concentrations to cause damage [45]. Individual spacesuits will contact during spacesuit removal when the spacesuit may be need to be designed to reduce the retention time of dust and to coated with the acid. Furthermore, airborne acid aerosols and prevent the build-up of high levels of dust in susceptible areas vapours in the airlock atmosphere that have failed to be effec- of the spacesuit for example, between the arm interlocks [46]. tively extracted may contribute to undetected spacesuit cor- Wind velocities of between 15 m/s to 32 m/s can occur rosion if they are not well maintained and cleaned effectively with saltation dust [47]. Saltation is a specific type of particle/ following removal and prior to donning. dust transported by fluids such as winds or water. There will be aeolian saltation on Mars during many of the storms [48]. 4.4 Dust Adhesion Astronauts and others will need to continuously monitor the strength of the winds whilst undertaking EVA in particular A number of mechanisms cause dust adhesion on Martian sur- the level of dust particles lifted off the Martian surface. The faces including covalent bonding and cementation. The various lifted dust could be accelerated downwind, then collide with processes affect the adhesiveness of the dust differently for -ex the ground to eject new dust particles and subsequently cause ample, in the size and concentration of the particles produced.

164 Vol 72 No.5 May 2019 JBIS MARTIAN DUST: the Formation, Composition, Toxicology, Astronaut Exposure Health Risks and Measures to Mitigate Exposure

Covalent bonding of dust particles or “sintering” should not follow any exposure to a hazard such as the Martian dust. The be a significant form of adhesion on Mars. However, particle severity of the health effects will be mainly related to the toxicity concentration and the subsequent adhesion on surfaces may of the dust and the routes of exposure. Due to incomplete data be expected to be significant for example, if there is adhesion to on the nature of the Martian dust, when judging the potential spacesuits [54], on spacecraft equipment and on mining tools health risks, an individual astronaut may react differently to the where there may be the potential for damage and loss of func- exposure depending on an individuals’ immune response. This tion power. uncertainty needs to be evaluated when an assessment of the health risks and subsequent evaluation is made on the potential 4.5 Tribo-electric Charging risks to health from dust exposure and when determining the measures to mitigate exposure [62]. Significant sand movement on Mars could lead to a large redis- tribution of dust around the whole planet and in localised are- If any of the astronauts in a spacecraft or living in a Martian as around a settlement. The subsequent impaction, deposition settlement develop disease from exposure to toxic dust via in- and rebound of the dust could cause tribo-electric fields suffi- halation, ingestion, skin absorption or by eye contact, then it cient to damage electrical devices that are necessary to ensure will be the space physician who will treat the symptoms of each a safe environment inside and outside of a settlement module. astronaut individually. In contrast, the astronautical hygienist Tribo-electric charging is a type of electric charge which can will identify the specific exposure health hazards and will pro- be generated by contact between materials through friction. vide an accurate characterisation of for example, the Martian The electric field caused by the tribo-electric charging on Mars dust to assess the potential health risks for all the astronauts could change the trajectories of sand and dust. The subsequent likely to be exposed [63]. The assessment will determine the resultant sand movement from the formation of dust devils most effective methods to be implemented to mitigate expo- could shape the landscape in particular if strong electrostatic sure. The implementation of the mitigation measures should fields were formed inside the dust devils [55]. The level of elec- ensure that the health of all the astronauts and others living and trification of the Martian dust devils and the affects on future working in the spacecraft or settlement is protected as well as EVAs, especially if there was damaged electrical equipment, that of an ill individual [1]. would need to be continuously monitored and evaluated in a risk assessment. Guitton [64] has argued that in the past, the application of space medicine has been mistaken as astronautical hygiene Tribo-electric charging or the effects of contact electrifi- (i.e. science and technology applied to control exposure in low cation and frictional charging could have a major impact on gravity environments) to the detriment of an astronaut’s health. spacecraft surfaces but the effects will vary depending on the Space medicine is human central and is involved with treat- mineralogical composition of the dust in particular the particle ing and screening an individually sick person. Astronautical size and shape. Turbulent Martian dust storms could produce 5 hygiene identifies the specific hazards encountered in space, KV/m electrical fields [56]. The high-level dust adhesion forces such as Martian dust, that could affect a specific population therefore as found on Mars, and due to the highly charged par- and therefore designs and implements the measures to protect ticles, could prevent dust being easily removed from spacecraft the whole of a potentially exposed population. To reduce the surfaces by electrical, vibrational and mechanical dust remov- health risks from dust exposure, both of these space disciplines al techniques [57]. However, dust particles could be removed should be utilised for the protection of the health of all space by a number of other techniques including lift off, sliding and workers including the astronauts [65]. To achieve this, in the rolling. These are the main removal methods and their effec- future, there will be a need for a paradigm shift to ensure that tiveness would vary depending on the levels of dust settled on each of the disciplines i.e. space medicine and astronautical surfaces [58]. It may be that the physical covering of surfaces hygiene have an understanding of how each can be effectively during a dust storm is the most effective method to prevent applied [66]. contact with specific equipment followed by the cleaning of contaminated surfaces by mechanical means. 5.2 Dust Particle Size and Toxicity

5 MARTIAN DUST AND TOXICOLOGY Martian dust is uniformly distributed over the whole planet. It is abrasive, electrostatic, magnetic, highly oxidative and chemi- 5.1 Introduction cally active. These combined properties contribute to the overall toxicity of the dust. Recurrent dust storms on the planet and Reports from the Apollo missions (i.e. six missions and twelve the formation of dust devils contribute to the uniformity of the astronaut EVAs) indicated that inhalation exposure to lunar dust by spreading it over the planet. dust resulted in cases of ill health. The Apollo astronaut dust exposures responsible for the ill health effects were uncontrolled, According to Perko et al. [15], the average diameter of the short-term and occurred following the removal of their space- Martian dust is between 0.06 µm–3.2 µm for the suspended suits. The exposures were sufficient to cause acute symptoms airborne dust, up to 100 µm for sand particles and up to 2 mm such as coughing, sore throat and shortness of breath [3] [59]. for agglomerates. Dust particle samples from around the “Vi- It is expected that similar acute symptoms of ill health will oc- king” and “Pathfinder” Landers have varied in size from 2.0 cur following the inhalation of Martian dust during spacesuit µm–10 µm and from 30.0 µm–40.0 µm respectively. The dust donning and doffing unless the exposure is efficiently and -effec that is < 10 µm is in the respirable range and there is the risk tively mitigated [60]. So, prior to and following EVA, the astro- of serious health effects if the dust is inhaled. Furthermore, ul- nautical hygienist will need to assess the dust exposure health trafine particles of <0.1 µm in the Martian dust, when inhaled risks based on the physical and chemical characteristics of the in significant levels, may cause oedema and lung inflammation Martian regolith/dust of the surrounding exploratory area [61]. if significant levels are inhaled [67]. Studies by Prisk [68] have shown that the size of the dust particles in a microgravity envi- Risk is the probability of an adverse health effect that will ronment, in particular inside a spacecraft, may affect the depo-

JBIS Vol 72 No.5 May 2019 165 JOHN CAIN sition of dust within the lung. The deeper the penetration and The types of symptoms that may arise from acute and chron- subsequent deposition of dust within the lung will relate to a ic exposure to the highly toxic substances in the Martian dust higher risk of respiratory ill health effects. are likely to be similar to those symptoms that occur in terrestrial workers [73]. However, the symptoms of acute exposure Photometric (measurement of light intensity) and pola- and the subsequent risks to health may be more common on rimetric (measurement of mainly radio and light waves) ob- Mars due to the nature of the work tasks, on the high levels of servations of Mars have been carried out by Morozenko and dust encountered during EVA and on the removal of contam- Vidmachenko [69] to categorise an object and to study particle inated space suits following EVA. There will also be a need for characteristics in the Martian atmosphere. They have shown prompt action in the event of a failure to prevent or control that particles in the atmosphere vary in size depending on exposure to the dust in particular during vehicle maintenance whether they are flattened, spheroid or spherical. Some par- procedures inside an airlock. ticles may double in size and hence if inhaled they will be deposited into the lower regions of the lung; sizes may range from The possibility of high peak exposures to the dust during 1.54 µm–1.62 µm. In the early stages of a dust storm particle some tasks (e.g. exhaust filter cleaning, spacesuit maintenance) sizes will range from between 1 µm and 20 µm and up to 8 µm– could result in significant synergistic and/or additive health ef- 10 µm during the period of greatest dust activity. In the final fects depending on the specific composition of the dust and the stages of a dust storm particle sizes will be approximately 1 µm reactions with other chemicals in the atmosphere. in size. The particle sizes measured indicate that residue dust inhaled during spacesuit removal in the airlock for example, Table 2 lists the most common hazardous substances found would reach the deep-seated parts of the lung. The acute inha- in Martian dust and the associated exposure health effects. lation exposure of such small particles or ultrafines i.e. <1.0µm could result in oedema, inflammation, fibrosis and possibly The severity of disease caused by exposure to the Martian cancer of the lung if exposure was chronic [37] [70]. dust will be related to the: – concentration of the toxic metal in the dust; Furthermore, during inhalation, the Martian dust particles – level of exposure dose; in particular those of nanoparticle size (<1µm) will be depos- – routes of exposure i.e. respiratory, dermal, ingestion, eye; ited in the nasal tracheo-bronchiole and alveolar regions of – toxicity of the substance the lung. Once deposited, these micro-sized particles may be – length of time of exposure; transported to the central nervous system. Krisanova et al. [71] – work tasks being undertaken; have found that once in the central nervous system there may – s usceptibility of the astronauts and other workers includ- be deleterious effects on glutamate homeostasis that could re- ing their gender; and sult in damage to the nervous system with symptoms of muscle – the measures used to mitigate dust exposure. weakness and damage to nerve endings. Martian dust contains a high level of iron oxides. Acute and 5.3 Dust Chemical Composition and Ill Health Effects chronic inhalation exposure to the iron oxides may result in a benign lung disease known as siderosis. Martian dust is known to consist of several highly toxic metallic compounds for example, silica dioxide, hexavalent chro- Overall, Martian dust is oxidative and caustic so exposure to mium and arsenic in varying concentrations. The dust also the skin and eye cornea may cause burning. Almadli et al. [74] contains perchlorates (ClO4-) which are associated with cal- have indicated that if an astronaut’s skin is damaged and there cium (Ca(ClO4)2) and magnesium chloride (Mg(ClO4)2) salts is exposure to Martian dust, then there is the potential that due in the dust. Individually, the concentrations of the hazardous to the highly oxidative nature of the dust that there may be in- substances that compose the dust will be low and not likely to creased cutaneous inflammation. If untreated then this could cause significant health effects following an acute exposure. lead to further skin damage and subsequent infection. Any However, exposure to known carcinogens such as silicates and monitoring of the skin and contaminated surfaces will also be cadmium at any concentration and at any exposure may over used to assess the health risks from ingestion of dust on contact a long latency period (i.e. chronic exposure) be significant and i.e. finger to mouth. cause specific diseases such as lung cancer and cardiovascular disease respectively [72]. The measures to mitigate exposure to the dust need to be

TABLE 2 The most common chemicals in Martian dust and the terrestrial and potentially extra-terrestrial exposure ill health effects Toxic metal and perchlorates in Martian dust Exposure ill health effects in terrestrial and potential extra-terrestrial settlement populations

1. Silicates/Respirable crystalline silica Chronic obstructive pulmonary disease (COPD), silicosis, lung disease e.g. cancer [73] 2. Hexavalent chromium Lung cancer, nose ulcers, pneumoconiosis characterised by coughing and fibrosis, COPD and pulmonary oedema [80] 3. Arsenic Skin, lung, liver, bladder and kidney cancer, ingestion can lead to death from acute exposure e.g. symptoms of vomiting, diarrhoea, gastric pain [72] 4. Cadmium Lung cancer, cardiovascular disease, hypertension, kidney disease, disrupts endocrine system [72] 5. Beryllium Lung cancer, chemical pneumonitis, granuloma of the lungs [81] 6. Perchlorates Blocks thyroid function by failing to absorb iodine, in children hypo-thyroidism can cause mental impairment [78]

166 Vol 72 No.5 May 2019 JBIS MARTIAN DUST: the Formation, Composition, Toxicology, Astronaut Exposure Health Risks and Measures to Mitigate Exposure stringent and sustainable hence the need for an effective per- of the toxicity of these chemicals following exposure [80]. A missible exposure limit for exposure to the dust. similar limit will be needed to control exposure to Martian dust and thereby reduce the risks to an astronaut’s health. However, 5.4 Perchlorates – Significant Health Hazards? until more exposure and toxicity data has been collected that is based on the characteristics of the Martian dust, the initial- Observations by Keller et al. [75] using Gamma Ray Spectrome- ly proposed permissible exposure limit (PEL) for exposure to try indicate that chlorine is widespread on the planet and there- the dust could be to the measurement of the dust as a whole. fore there is the potential for perchlorates to form. The Phoenix Otherwise, as a major component in dust, the PEL could be for Mars Lander has detected perchlorates in the Martian regolith the measurement of respirable crystalline silica (RCS) only as a at concentrations of 0.4 to 0.6 wt/% [76]. The perchlorates are surrogate marker of exposure. composed of the chlorine salts Ca (ClO4)2 and/or Mg (ClO4)2. Significant levels of the salts and perchlorates were first observed The reasons for choosing RCS as a surrogate marker of ex- on the surface of Mars in 2008 [77]. The Martian dust may con- posure to Martian dust rather than to one of the other com- tain up to 1% perchlorate [78]. Acute exposure to such concen- ponent toxic chemicals is because the presence of silicates trations would significantly increase an astronaut’s health risks. is widespread in the surface dust and they can be accurately measured in the dust samples. It is expected that any PEL that Spacesuit exposure to perchlorates will be difficult to avoid controls exposure to silicates in the dust will also control expo- during EVA in particular if perchlorates are mined for use in sure to the other toxic chemical constituents that are present fuel surface operations and to act as a source of oxygen. Direct in the dust albeit in much lower concentrations. Furthermore, inhalation of a few milligrams of dust that contain perchlorates the proposed PEL should also reduce the risks of health effects during spacesuit removal for example, could result in fast ab- associated with ingestion, eye contact and skin damage in par- sorption of the chemical inside the body. Together with skin ticular if there is sustained compliance with the PEL. absorption and ingestion, exposure to perchlorates could result in subsequent significant systemic health effects. Cain [2] has It would be impractical to set a PEL for Martian dust based indicated that ingestion of perchlorate in contaminated food on the lowest concentration of a toxic substance in the dust for and water could increase the body burden and impair the nor- example, that of beryllium (e.g. in Great Britain the assigned mal function of the thyroid resulting in symptoms of lethargy exposure limit for beryllium is 0.002 mg/m³ 8-hour Time and weight loss. Weighted Average) [81]. This is because of the efficiency and effectiveness of the mitigation systems that would be required Perchlorate salts are oxidative and can be absorbed through to ensure compliance with the proposed assigned PEL together the skin. If the levels of dermal exposure are high and uncon- with the economic costs of installing and maintaining any me- trolled then the thyroid could be seriously affected. Eye contact chanical controls. with perchlorates and associated dust could result in eye irritation with symptoms of reddening and itchiness. The toxicological evidence gathered and the astronautical hygiene data collected so far on which to set a proposed PEL A basic understanding of the mechanisms for perchlorate/ for Martian dust is mainly based on studies of simulated Mar- perchloric acid formation is necessary for assessing the overall tian and lunar dust studies [82] and on data from the Apollo exposure health risks and to determine the measures necessary missions [83]. Using RCS as a surrogate for Martian dust expo- to mitigate exposure. A risk assessment will need to evaluate sure indicates that a PEL of 0.05 mg/m³ (aerodynamic particles the concentrations of perchlorates in a particular area prior to size of 0.1 µm to 10µm) should be recommended for person- EVA as they may change over time in response to weather con- al short-term exposure to RCS of up to 3-4 hours. This level ditions and as yet unknown surface chemistry. The concentra- would account for the acute inhalation exposures that might be tions of perchlorates around a settlement, therefore, will need expected in the airlock during the removal of a spacesuit. The frequent monitoring to determine the levels in the dust and to PEL is a mass-based recommendation which will be independ- take appropriate action if the levels are likely to be high and ent of particle sizes in the 0.1 µm–10 µm range [84]. could lead to potential harm if exposure was uncontrolled. Long-term exposure to Martian dust from 1-3 months, 3 – 6 Wilson et al. [77] have proposed that Martian perchlorates months and from 6-12 months will each require an individual are formed by the radiolysis of the Martian surface due to ga- PEL for example, to control airborne dust as may arise during lactic cosmic rays together with the sublimation of chlorine ox- frequent EVA in mining areas and with the subsequent donning ides into the atmosphere and their subsequent synthesis. This and doffing of a spacesuit during maintenance work. The long is unlike other chemical formation mechanisms in the Solar term PEL(s) proposed will be related to the specific type of work System. Perchloric acid (HClO4) formed in the atmosphere being undertaken, on the likely routes of dust exposure and on will be deposited on the surface of the planet with the subse- the length of time of the work involved. Both the short and long- quent mineralisation of HClO4 in the regolith to from surface term PEL will need to account for the combined airborne dust perchlorates. Additional data from atmospheric and volcanic load including microbes in the air, particles arising from day-to- studies should provide further evidence on which to clarify the day working and living in low gravity environments [85]. proposed formation. However, the injection of chlorine into the atmosphere to produce perchlorates could be due to vul- When setting a PEL for Martian dust and to ensure compli- canisation and not by radiolysis [79]. ance to the PEL, the level set will need to be achievable, practical and sustainable over the allocated time period. It will need 5.5 Setting of a Permissible Exposure Limit for Martian Dust to reflect both the potential acute and chronic exposure health risks. The PEL set will need to be regularly revised to account Terrestrially, occupational exposure limits (OELs) have been for new data from toxicology studies, from astronautical hy- assigned for individual chemicals such as arsenic, hydrazine giene exposure studies and from as yet any unknown factors (used in rocket propellant) and hexavalent chromium because that may be found to affect exposure. The type and implemen-

JBIS Vol 72 No.5 May 2019 167 JOHN CAIN tation of effective airborne dust mitigation methods will be de- sults of other monitoring regimes could be used to judge the pendent on the length of time of dust exposure and the likely spatial and temporal variability of the environment and there- potential airborne concentrations as determined by the risk as- by gain an understanding of the potential for where high dust sessment. For example, the control used to comply with a PEL exposures could occur [88]. Computer modelling will be useful of 0.05 mg/m³ for up to 3-4 hours potential exposure could be for producing a statistical framework to quantify any uncer- the wearing of a disposable respirator with maximum particle tainty in the spatial and temporally varying data [89]. filtration during routine tasks such as the changing of filters in extraction systems. Furthermore, once dust monitoring has been undertaken then the use of hazard mapping could be applied to judge an In contrast to PELs, Spacecraft Maximum Allowable Con- astronautical hygiene risk assessment, to decide on how the risk centrations (SMACs) have been set for hundreds of specific air- assessment should be communicated and to outline the ways borne chemicals found in spacecraft such as aldehydes, xylene forward. The hazard maps will be useful for designing hazard in- and toluene. The exposure periods range from 1 hour to 180 tervention and control strategies for Martian dust exposure [90]. days to reflect the potential exposure and control. The exposure levels set are designed to protect the astronaut if exposures 6 MEASURES TO MITIGATE EXPOSURE TO MARTIAN DUST are at chemical airborne concentrations below the appropriate SMAC [60]. Over time, it is likely that specific SMACs will be 6.1 Dust Exposure Mechanical Mitigation Methods set for airborne chemicals in the Martian settlements. Effective astronautical hygiene principles should be applied to 5.6 Monitoring Strategies to Assess the Exposure Health prevent or control exposure to Martian dust and thereby sig- Risks nificantly reduce the dust exposure health risks from potential inhalation, dermal, ingestion and/or eye contact. These meas- Random personal inhalation exposure monitoring for RCS will ures will include: need to be undertaken at appropriate times and places when – Th e elimination of the dust inside the airlock and the set- the spacesuit dust burden is likely to be high and constitute tlement environment using appropriate techniques e.g. the a risk to health such as in an airlock following deep ground use of an electrostatic precipitator to remove dust from the mining and the subsequent cleaning of drilling and other tools. Martian air during the extraction of oxygen, fuel and water Such monitoring would ensure that the 3-4 hours short-term from the Martian air [91], the provision of effective high exposure PEL of 0.05 mg/m³ was being complied with. In ad- recirculation rate air removal systems with high speed fans dition, an installed alarm system in an appropriate place would and high efficiency particulate or HEPA filters [92]; register any high exposure levels when they occurred. Contin- uous background airborne concentration monitoring would – The prevention or significant reduction of the airborne need to be undertaken to measure the surrounding airborne dust throughout the spacecraft, on settlement modules, Martian dust levels in for example, a Rover where exposure was on the EVA Rovers and during mining e.g. by the use of evident but spacesuit helmets didn’t need to be continuously suitable electrodynamic dust shields [93]; worn. Any significant fluctuations in the level of exposure i.e. above the PEL could pose a high risk. Immediate action would – Th e use of suitable spacesuit materials with low dust re- therefore be necessary, the causes investigated e.g. a blocked tention, the use of effective adhesive reduction methods filter with potential dust blow back and prompt action taken. by surface modification [94],

Random background airborne concentration monitoring to – Using effective dust removal techniques such as appro- measure RCS and other hazardous substances including mi- priate water dust suppression procedures in the airlock crobes would also need to be undertaken as part of a general during spacesuit removal, by the cleaning of equipment monitoring regime to check compliance with the PEL. If high by using vacuums and other methods to remove localised levels of dust were measured above the PEL, then remedial ac- dust from spacesuits and subsequently clean them [95]; tion would be needed to identify the fault and immediate action taken. – P reventing the build-up of dust on equipment and surfaces during EVA by designing robust equipment that have The terrestrial monitoring and analytical techniques to few areas where dust could accumulate and cause damage; measure airborne RCS will need to be adapted for the measurement of Martian dust and the necessary monitoring design – P roviding laboratory glove boxes (e.g. for biological, life strategies developed. The use of a portable system to measure science, portable and microgravity science use) with ef- direct dust exposure in specific environments may be appro- fective containment, filtration systems and air extraction priate and/or the wearing of a personal monitor by astronauts airflow speeds. when they are on EVA [86]. – En suring that during operation, the glove box work vol- In addition to airborne exposure monitoring for toxic sub- ume is at a constant pressure of 3-10 mbar relative to the stances, there may be a need to measure skin exposure by direct main laboratory environment to ensure that any contam- skin sampling. Surface dust level contamination may also need inated material remains inside the glove box in the event to be monitored especially in the airlock where contaminated of a leak [96]; and spacesuits are removed following EVA and to check on the effi- cacy of exhaust ventilation systems and cleaning regimes where – Wearing suitable personal protective equipment to pre- appropriate [87]. vent inhalation exposure to dust during work on short- term tasks such as maintenance of local exhaust ventila- The results from the personal exposure monitoring for RCS, tion equipment where localised dust exposures may be from the continuous direct background sampling and the re- high and PELs may be breached [97].

168 Vol 72 No.5 May 2019 JBIS MARTIAN DUST: the Formation, Composition, Toxicology, Astronaut Exposure Health Risks and Measures to Mitigate Exposure

6.2 Spacesuits ous substances that compose the dust such as silicates, hexavalent chromium and arsenic. There will be the potential for During EVA, the main means to mitigate astronaut dust ex- symptoms of ill health if the dust is inhaled, ingested, absorbed posure and to protect against other extreme environmental through the skin or is in contact with the eye. A PEL for Mar- hazards such as radiation and freezing temperatures will be by tian dust of 0.05 mg/m3 (aerodynamic particle size of 0.1 um wearing a spacesuit with low dust retention properties. To pro- to 10 um) has been proposed to control exposure to dust and tect the astronaut from heavy dust exposure and the potential thereby to reduce the exposure health risks. If the PEL is ex- for adhesion of the dust onto the spacesuit, the level of protec- ceeded for example, in an airlock during spacesuit removal tion provided by an individual spacesuit worn will depend on then exposure control will not have been achieved. The reasons several factors. These factors will include the: why the failure has occurred will need to be investigated and – nature of the work to be undertaken on the extreme Mar- promptly remedied. tian surface; – time period of the work; Chlorine is widespread on Mars and is responsible for per- – types of terrain encountered; chlorate formation. The health risks associated with exposure – composition of the dust; to the perchlorates are serious i.e. they impair the proper func- – surrounding levels of dust; tioning of the thyroid gland that could result in lethargy, goitre – potential to coat equipment; and and reduced metabolic rates. The concentrations of the per- – potential for the dust to become airborne when the space- chlorates may change on the Martian surface in response to the suit is removed in the airlock. weather conditions and the formation of dust storms and dust devils. The higher perchlorate concentration areas could in- The individual spacesuit chosen to be worn during a specific crease the exposure health risks unless exposure was prevented EVA or other tasks will take account of the above factors and or controlled by appropriate measures. the judgement and experience of the astronautical hygienist who will choose the spacesuit. Other factors could significantly increase the astronaut risks to health during EVA. These include the formation of acid fog In addition to the above factors, the spacesuit will need to with the potential to damage spacesuits over time and surface be flexible and ergonomically designed to ensure that muscu- sand movements with a heavy build-up of dust of different loskeletal health problems are minimised and that if necessary particle sizes that could adhere to the spacesuit. Dust arising can provide protection from the extreme conditions on Mars from dust devils, dust storms and near surface saltation togeth- for up to 8 hours and longer if required. To obtain these proper with dust adhesion on surfaces and tribo-electric charging erties, component parts of individual spacesuits may need to could increase the health risks when astronaut spacesuits are be interchangeable. coated with the dust and removed in an airlock.

Particular importance for each spacesuit will be the need The application of the principles of astronautical hygiene to provide a life support system that generates oxygen and should be used to mitigate Martian dust exposure [1] for ex- removes carbon dioxide together with ensuring that cold sur- ample, by the wearing of individually designed spacesuits that face temperatures don’t make the spacesuit brittle and difficult are flexible and ergonomically usable and by utilising the prop- to function. The correct choice of materials that make up the erties of the dust e.g. by the use of an electrostatic precipitator spacesuit will therefore be essential. to remove air from the internal environment and by utilising surface adhesion modifiers to reduce dust retention. Marcy et al. [46] and Gallini [98], tested various materials for use in a Martian spacesuit (e.g. Kevlar to protect from Future research work in this field will include: impact, puncture and fire, polyimides to provide a structured silicon elastomer to ensure a dampening layer). They conclud- – p roviding more data on the characteristics of the Martian ed that three multi-layered functioning garments consisting dust from various sites on the planet and use this data of a liquid cooling ventilation garment, a pressure suit and a to provide a better understanding of their relationship to thermomechanical garment that was based on layers of differ- potential exposure ill health effects [100]; ent materials would suffice and thereby protect the astronaut from the hostile Martian environment. Such a layered structure – un dertaking more studies on which to base a more accu- would also ensure effective pressurisation, provide an integrat- rate PEL for example, by computer modelling, by animal ed life support system and ensure flexibility [99]. The design studies, by Martian dust simulation tests, by understand- and development of prototype spacesuits will be essential when ing the toxicological effects of exposure and by undertak- deciding on the most effective design parameters. ing monitoring surveys of the astronauts during specific tasks; 7 CONCLUSIONS – de veloping more effective dust exposure risk assessment This paper discusses the geology of Mars and the characteris- models based on real inhalation airborne exposure mon- tics of the Martian dust as they relate to exposure toxicity and itoring data to determine and apply the most appropriate the risks to astronaut health. These factors are all significant for sustainable measures to mitigate dust exposure [101]; determining the most appropriate measures to mitigate astronaut dust exposure following a risk assessment. – t he designing of spacesuits that are suitable for providing adequate protection during all types of EVA in extreme Exposure to Martian dust may cause acute or chronic health Martian environments; and effects e.g. respiratory irritation and lung cancer respectively. The exposure health effects will be related to the chemical – de veloping techniques to provide emergency pressuriza- composition of the dust and the concentration of the hazard- tion of the spacesuit if it is damaged during EVA.

JBIS Vol 72 No.5 May 2019 169 JOHN CAIN

REFERENCES 1. J. R. Cain, “Astronautical hygiene – a new discipline to protect the 28. F. Goesmann, W. B. Brinckerhoff, F. Raulin et al., “The Mars Organic health of astronauts working in space.” JBIS, 64, pp.179–185, 2011. Molecule Analyzer (MOMA) Instrument: Characterization of Organic 2. J. R. Cain, “Humans in space and chemical risks to health.” SpaceFlight, Material in Martian Sediments.” Astrobiology, 17, pp. 655 – 685, 2017. 58, pp. 336–341, 2016. 29. J. L. Bishop, S. L. Murchie, C. M. Pieters, A. P. Zent, “A model for 3. J. R. Cain, “Lunar dust: the hazard and astronaut exposure risks.” Earth, formation of dust, soil, and rock coatings on Mars: Physical and chemical Moon, Planets, 107, pp. 107–125, 2010. processes.” Journal of Geophysical Research, 107, pp. 1-17, 2002. 4. D. Linnarsson, J. Carpenter, B. Fubini et al., “Toxicity of Lunar dust.” 30. C. N. Foley, T. E. Economou, R. N. Clayton, “Chemistry of Mars Planet Space Sci., 74, pp. 57–71, 2012. Pathfinder samples determined by the APXS.” Lunar Planet Si. XXX11, abstract 1979, 2001. 5. Y. Nakamura, G. V. Latham, H. J. Dorman, “Apollo lunar seismic experiment – final summary.” Journal of Geophysical Research, 87, pp. 31. H.Y. McSween et al., “Chemical multispectral and textural constraints A117–A123, 1982. on the composition and origin of rocks at the Mars Pathfinder landing site.” J. Geophys. Res., 104, pp. 8679–9715, 1999. 6. J. F. Bell et al., “Minerologic and compositional properties of Martian soil and dust: Results from Mars Pathfinder.” J. Geophys. Res., 105, pp. 32. H. Y. McSween, K. Keil, “Mixing relationships in the Martian regolith 1721–1755, 2000. and the composition of the globally homogenous dust.” Geo-chem. Cosmochim. Acta., 64, pp. 2155–2166, 2000. 7. H. Y. McSween, G. J. Taylor, M. B. Wyatt, “Elemental composition of the Martian crust.” Science, 324, pp. 736–739, 2009. 33. J. A. Crisp, “The effect of thin coatings of dust or soil on the bulk APXS composition of the underlying rocks at the Pathfinder landing site.” 8. M. T. Zuber, “The crust and mantle of Mars.” Nature, 412, pp. 220–227, Lunar Planet Sci. XX1X, abstract 1962, 1998. 2000. 34. S. Murchie, O. Barmouin – Jua et al., “Diverse rock types at the Mars 9. J. R. Cain, “Respirable crystalline silica – a failure to control exposure.” Pathfinder landing site.” Lunar Planet Sci. XXX1, abstract 1267, 2000b. J. Phys. Conf. Ser., 151 2009 doi: 10.1088/1742-6596/151/1/012010. 35. J. F. Mustard, C. D. Cooper, M. K. Rifkin, “Evidence for recent climate 10. R. E. Arvidson, J. L. Gooding, H. L. Moore, “The Martian surface as change on Mars from the identification of youthful near-surface ground imaged, sampled and analysed by the Viking landers.” Rev. Geophys., 27, ice.” Nature, 412, pp. 411–414, 2001. pp. 39–60, 1989. 36. A. P. Zent, “On the thickness of the oxidized layer of the Martian 11. C. Vita-Finzi, Planetary Geology – An Introduction, Terra Publishing, regolith.” J. Geophys. Res., 103, pp. 31491–1498, 1998. Harpenden, pp. 25– 7, 2005. 37. K. Donaldson, V. Stone, “Current hypotheses on the mechanisms of 12. S. L. Brantley, N. P. Mellott, “Surface area and porosity of primary toxicity of ultrafine particles.” Ann. Ist. Super. Sanita., 39, pp. 405–410, silicate minerals”. Am. Miner., 85, pp. 1767–1783, 2015. 2003. 13. L. Maltagliati et al., “Evidence of water vapour in excess of saturation in 38. J. W. Head, “The Geology of Mars: New insights and outstanding the atmosphere of Mars.” Science, 33, pp. 868–1871, 2011. questions.” In: The Geology of Mars: Evidence from Earth-Based 14. W. V. Baynton et al., “Distribution of hydrogen in the near surface of Analogues (ed Mary Chapman), Cambridge University Press, 2007. Mars. Evidence for sub-surface ice deposits”. Science, 297, 81–85, 2002. 39. P. R. Christensen, J. L. Bandfield, M. D. Smith, V. E. Hamilton, R. N. 15. H. A. Perko, J. D. Nelson, J.R. Greene, “Review of Martian Dust: Clark, “Identification of a basaltic component on the Martian surface Composition, Transport, Deposition, Adhesion and Removal”, in from thermal emission spectrometer data.” Journal of Geophysical Proceedings of Engineering, Infrastructure and Science in Space, ASCE Research (Planets)., 105, pp. 9609–9622, 2000a. Press, 2002. 40. P. R. Christensen, J. L. Bandfield, et al., “Detection of crystalline 16. G. Certini, R. Scalenghe, R. Amundson, “A view of extra-terrestrial hematite mineralization on Mars by the thermal emission spectrometer: soils.” European Journal Soil Science, 60, pp.107–1092, 2009. evidence for near – surface water.” Journal of Geophysical Reseach 17. S. M. McLennan, “Chemical composition of Martian soil and rocks: (Planets), 105, pp. 9623, 2000b. Complex mixing and sedimentary transport.” Geophy. Res. Letts., 27, pp. 41. J. W. Head, L. Wilson, “Mars: a review and synthesis of general 1335–1338, 2000. environments and geological settings of magma – H20 interactions.” 18. J. B. Pollack, D. S. Colburn, F. M. Flasar et al., “Properties and effects of Geological Society Special Publication, 202, pp. 27–57, 2002. dust particles suspended in the Martian atmosphere.” J. Geophys. Res., 42. P. Masson, M. H. Carr, F. Costard et al., Geomorphologic evidence for 84, pp. 2929–2945, 1979. liquid water. In: Chronology and evolution of Mars (ed R. Kallenbach, 19. R. Greeley, S. H. Williams, “Dust deposits on Mars – the Parna analog.” J. Giess, W.K. Hartmann). Dordrect: Kluver Academic, pp. 333–364, Icarus, 110, pp. 165–177, 1994. 2001. 20. T. Ding, K. Zhang et al., “Simulation of adhesion characteristics of 43. K. P. Harrison, R. E. Grimm, “Rheological constraints on Martian Martian dust in the atmospheric environment of Mars based on Edem- landslides.” Icarus, 163, pp. 347–362 2003. Fluent.” Journal of Applied Physical Science International, 5, pp. 241–249, 44. J. Appelbaum, G. A. Landis, I. Sherman, “Solar radiation on Mars – 2016. update 1991.” Solar Energy, 50, pp. 35–51, 1993. 21. M. A. Kahre, J. R. Murphy, R. M. Haberle, “Modelling the Martian 45. G. J. Roebelen, Spacesuit cooling apparatus. US Patent No 5092, 129, dust cycle and surface dust reservoirs with the NASA Ames general March 3rd 1992. circulation model.” J. Geophys. Res., 111, pp. E06008 2006. 46. J. L. Marcy, A. C. Shalanski et al., “Material choices for Mars.” Journal of 22. Y. Liu, C. S. Cockell, G. Wang et al., “Control of Lunar and Martian dust Materials Engineering and Performance, 13, pp. 208–217 2004. – Experimental Insights from Artificial and Natural Cyanobacterial and 47. J. R. Gaier, M. E. Perez-Davis, M. Marabito, Aeolian Removal of dust Algal crusts in the Desert of Inner Mongolia, China.” Astrobiology, 8, from Photovoltaic Surfaces on Mars. NASA Technical Memorandum, pp. 75 – 86, 2008. 102 507 1990a, 1990. 23. G. A. Landis, P. P. Jenkins, “Dust Accumulation and Removal 48. M. P. Almeida, E. J. R. Parteli, J. S.Andrade Jr, H.J. Hermann, “Giant Technology (DART) Experiment on the Mars 2001 Surveyor or Lander”, saltations on Mars.” PNA, 105, pp. 6222–6226 2008. Proceedings of the Second World Conference on Photovoltaic Energy Conversion, 111, pp. 3699 – 3702, 1998. 49. R. O. Kuzmin, R. Greeley, R. Rafkin et al., “Wind-related modification of some small impact craters on Mars.” Icarus, 153, pp. 61–70, 2001. 24. G. J. Flynn, D. S. McKay, “An assessment of the meteoritic contribution to the martian soil.” J. Geophys. Res Solid Earth, 95, pp. 14497–14509, 50. P. B. Niles, J. Michalski, D. W. Ming, D. C. Golden, “Elevated olivine 1990. weathering rates and sulfate formation at cryogenic temperatures on Mars.” Nature Communications, 8, pp. 1–5, 2017. 25. K. L. Thomas-Keprta, S. J. Clemett et al., “Origins of magnetite nanocrystals in Martian Meteor ALH84001.” Geochemica et 51. K. L. Tanaka, “Dust and ice deposition in the Martian geologic record.” Cosmochimica Acta., 73, pp. 6631–6677, 2009. Icarus, 144, pp. 254–266, 2000. 26. D. J. Stevensen, “Mars’ core and magnetism.” Nature, 412, pp. 214–219, 52. A. A. Olsen, J. D. Rimstidt, “Using a mineral lifetime diagram to evaluate 2001. the persistence of olivine on Mars.” Am. Min., 92, pp. 598 602, 2007. 27. W. V. Boynton, D. W. Ming, S. P. Kounaves et al., “Evidence for calcium 53. E. M. Hausrath, S. L. Brantley, “Basalt and olivine dissolution under carbonate at the Mars Phoenix landing site.” Science, 325, pp. 61–64, cold, salty and acidic conditions.” What can we learn about aqueous 2009. weathering on Mars? J. Geophys. Res., 115, pp. E12001, 2010.

170 Vol 72 No.5 May 2019 JBIS MARTIAN DUST: the Formation, Composition, Toxicology, Astronaut Exposure Health Risks and Measures to Mitigate Exposure

54. M. Reyders, B. B. Grosshauser, A. Smarandache et al., Effects of 78. A. F. Davila, D. Wilson, J. D. Coates, C. P. McKay, “Perchlorate on Mars: Lunar and Mars dust simulants on HaCaT keratinocytes and CHOK1 a chemical hazard and a resource for humans.” International Journal of fibroblasts. Adv. Space Res., 47, pp. 1200–1213, 2011. Astrobiology, pp. 1 – 5, 2013. 55. N. O. Reino, J. F. Kok, “Electrical activity and dust lifting on Earth, Mars 79. D. C. Catling, M. W. Claire, K. J. Zahnle et al., “Atmospheric origins and beyond.” Space Sci. Rev., 137, pp. 419–434, 2008. of perchlorate on Mars and in the Atacama.” J. Geophys. Res., 115, pp. 56. D. D. Sentman, “Electrostatic fields in a dusty Martian environment.” E00E11, 2010. Sand and Dust on Mars, NASA Conference Publication CP-10074, 53, 80. Health and Safety Executive, EH40/2005 Workplace exposure limits: 1991. Containing the list of workplace exposure limits for use with the Control 57. C. I. Calle, J. L. McFall, C. R. Buhler et al., Dust particle removal by of Substances Hazardous to Health Regulations 2002. Environment electrostatic and dielectrophoretic forces with application to NASA Hygiene Guidance Note EH40 HSE Books, 2005. exploration missions. Proceedings ESA Annual Meeting on Electrostatics 81. Health and Safety Executive, “Beryllium and you – working with 2008, Paper 01. beryllium – are you at risk?”, 2013, www.hse.gov.uk/pubns/indg311. pdf 58. J. R. Gaier, et al., Effects of dust accumulation and removal on radiator 82. C. W. Lam, J. T. James et al, “Pulmonary toxicity of simulated lunar and surfaces. NASA Technical Memorandum 103704, 1991. Martian dusts in mice: Histopathology 7 & 90 days after intratracheal 59. J. R. Cain, “Moon dust – a danger to lunar explorers.” SpaceFlight, 52, pp. instillation.” Inhal. Toxicol., 14, pp. 901 – 916, 2002. 60–65, 2010. 83. R. R. Scully, V. E. Meyers, Human Research Program Space Human Factors 60. J. T. James, D. E. Gardner, “Exposure limits for airborne contaminants and Habitability (SHFH) Element. NASA, Lyndon B Johnson Space in spacecraft atmospheres.”Appl. Occup. Environ. Hygiene, 11, pp. Center, Houston, Texas, 2015. 1424–1432, 1996. 84. P. S. Greenberg, D. R. Chen, S. A. Smith, Aerosol measurements of the 61. P. A. Sim, “Martian dust and its interaction with human physiology: an fine and ultrafine particle content of lunar regolith. NASA/TM-2007- emergency physicians approach.” Dust in the Atmosphere of Mars 2017, 214956 http://gltrs.grc.nasa.gov/ 2007. LPI contrib. No 1966, 85. J. Dory, Constellation program human-systems integration requirements; 62. S. O. Funtowicz, J. R. Ravetz, “Uncertainty, complexity and post – normal final report. NASA CxP 70024, 2006. science.” Environmental Toxicology and Chemistry, 13, pp. 1881–1885, 86. British Standards Institution 1993, Workplace Atmospheres - size fraction 1994. definitions for measurement of airborne particles. BS EN 481, 1993. 63. J. R. Cain, “Astronaut health – planetary exploration and the limitations 87. NASA, 2015 Dust Risk Review Panel – Research Plan Review for: on freedom,” In: The meaning of liberty beyond Earth, C. S. Cockell (Ed), The Risk of Adverse Health and Performance Effects of Celestial Dust New York, Springer, 2014. Exposure, NASA Technical Report, 2015. 64. J. Guitton, “Challenges in Space Medicine.” PHF, 1, pp. 73–77, 2012. 88. H. Chang, N. D. Le, A. Fu, J. V. Zidek, “Designing Environmental 65. S. M. Grenon, J. Saary, “Challenges in aerospace medicine education.” Monitoring Networks for Measuring extremes.” Environmental and Aviat. Space Environ. Med., 82, pp. 1071–1072, 2011. Ecological Statistics, 14, pp. 301–321, 200.) 66. B. J. O’Brien, “Paradigm shifts about dust on the Moon: from Apollo 11 89. D. Le Nhu, J. V. Zidek, Statistical Analysis of Environmental Space-Time to Chang’e – 4.” Planetary and Space Science., 156, pp. 47–56, 2018. Processes. Springer 2006. Springer–Verlag, New York, 2006. 67. K Donaldson, V Stone, A. Clouter, L. Renwick, W. MacNee, “Ultrafine 90. K. Kirsten, A. Koehler, J. Volkens, “Prospects and Pitfalls of Occupational particles.” Occup. Environ. Med., 588, pp. 211–216, 2001. Hazard Mapping: Between these lines there be dragons.” Annals of Occ. Hyg,. 55, pp. 829–840, 2011. 68. G. K. Prisk, “Microgravity and the respiratory system.” Eur Respir J., 43, pp. 1459–1471, 2014. 91. B. S. Williams, Dust Mitigation for Martian Exploration. NASA WRP– Internship Final Report, Kennedy Space Centre, 2011. 69. A. N. Morozhenko, A. P. Vidmachenko, “Optical parameters of Martian dust and its influence on the exploration of Mars.” LPI Contrib. No 1966, 92. J. H. Agui, D. P. Stoker, NASA lunar dust filtration and separations 2017. workshop report. NASA/TM-2009-215821, 2009. 70. D. J. Anderson, D. O. Wilson, A. Hirsch, “Respiratory tract deposition of 93. C. I. Calle, P. J. Mackey, M. D. Hogue et al., “Electrodynamic dust shields ultrafine particles in subject with obstructive or restrictive lung disease.” on the International Space Station: Exposure to the space environment.” Chest, 97, pp. 1115–1120, 1990. Journal of Electrostatic, 17, pp. 257–259, 2013. 71. N. Krisanova, L. Kasatkina, R. Sivco et al., “Neurotoxic potential of Lunar 94. A. Dove, G. Devaud, Xn Wang et al., “Mitigation of lunar dust adhesion and Martian dust: Influence on Em, Proton Gradient, Active Transport by surface modification.” Planetary and Space Science, 59, pp. 1784–1790, and binding of Glutamate in Rat Brain Nerve Terminals.” Astrobiology, 2011. 13, pp. 679 – 692, 2013. 95. P. E. Clark, S. A Curtis, F. A. Minetto, M. Moore, J. Nuth, “Characterizing 72. R. W. Clapp, M. M. Jacobs, E. L. Loechler, “Environmental and physical and electrostatic properties of lunar dust as a basis for Occupational Causes of Cancer New Evidence, 2005 – 2007.” Rev. developing dust removal tools.” NLSI Lunar Science Conference, 2008. Environ. Health, 23, pp. 1 – 37, 2008. 96. R. H. Kruse, W. H. Puckett, W. H. Richardson, “Biological safety 73. B. Pesch, M. Lehnert, T. Weiss et al., “Exposure to hexavalent chromium cabinetry.” Clin. Microbiol. Rev., 4, pp. 207-241, 1991. in welders: Results of the WELDOX 11 field study.” Annals of Work 97. J. T. James, Air Quality Standards for Space Vehicles and Habitats. SAE Exposure and Health, 62, pp. 351–361, 2018. Technical Paper 2008-01-2125, 2008. 74. G. Almadli, R. Schnabel, A. Jokuszies, et al., “Impact of Martian and 98. J. Gallini, “Polyamides, Aromatic.” In: Encyclopedia of Polymer Sciences Lunar dust simulants on cellular inflammation in human skin wounds ex and Technology, John Wiley & Sons, New York, 2002. v i v o.” Handchir Mikrochir Plast Chir., 46, pp. 361–368, 2014. 99. L. K. Massay, Permeability Properties of Plastics and Elastomers. Plastics 75. J. M. Keller, et al., “Equatorial and mid latitude distribution of chlorine Design Laboratory, Morris NY, 1999. measured by Mars Odyssey GRS.” J. Geophys. Res., 111, pp. E03508, 2006. 100. J. L. Bishop, S. L. Murchie, C. M. Pieters, A. P. Zent, “A model for 76. S. C. Cull, R.E. Anvidson, J. G. Catalaso, D. W. Ming et al., “Concentrated generation of Martian surface dust, soil and rock coatings: Physical vs. perchlorate at the Mars Phoenix landing site: evidence for thin film liquid Chemical interactions and palagonitic plus hydrothermal alteration.” water on Mars.” Geophys. Res. Lett., 37, pp. L22203, 2010. In 5th International Mars Conference, abstract 6222, Jet Propul. Lab. 77. E. H. Wilson, S. K. Atreya, R. I. Kaiser, P. F. Mahaffy, “Perchlorate Pasadena, California, 1999. f=ormation on Mars through surface radiolysis initiated atmospheric 101. M. M. Quinn, P. M. Smith, “Gender, Work and Health.” Annals of Work chemistry: a potential mechanism.” Journal of Geophysical Research: Exposures and Health, 62, pp. 389 – 392, 2018. Planets, 121, pp. 1472–1487, 2016.

Received 17 January 2019 Approved 3 April 2019

JBIS Vol 72 No.5 May 2019 171 JBIS VOLUME 71 2018 PAGES 172–179

THE CONCEPT OF A LOW COST SPACE MISSION to Measure “Big G”

ROGER LONGSTAFF, Guest Associates (Europe) Ltd., 2 Alyson Villas, Vincent Close, Ilford, United Kingdom email [email protected]

The Universal Gravitational constant “G” defines the force of gravity – one of the four fundamental forces of nature. However, the currently accepted value of G is orders of magnitude less accurate than the other fundamental constants, and ground based experiments differ from each other by as much as 500 ppm in its measurement. This paper describes a simple, space based experiment that could measure G to within 50 ppm, thereby producing a valuable scientific measurement using a completely different method and subject to different sources of error. The experiment could also contribute to the ongoing investigation into the nature of gravity.

Keywords: Universal Gravitational Constant, ISS experiment

1 INTRODUCTION cosmology and astrophysics. However, the known value of G is orders of magnitude less accurate than the other fundamen- The key science question the experiment addresses is the de- tal constants. The current best understanding of G is the CO- termination of the Universal Gravitational Constant – G – in DATA definition shown in Fig. 1. This shows in summary that pure, ballistic conditions. The chosen experiment would be per- the results of recent ground-based experiments differ in this formed in Earth orbit, aboard the International Space Station determination by over 500 ppm (even though some of them (ISS). The method is to measure the self-attraction of two test purport to be accurate to 20 ppm) and the physics community masses using laser metrology – the same technology that has in the USA has declared that the resolution of this anomaly is been flight qualified by the Lisa Pathfinder mission [1]. The ISS a national goal [2]. has been chosen for this experiment as it will enable a relatively low cost mission, with regular launch opportunities and astro- The science objective of the experiment is to achieve a meas- naut intervention to configure and operate the experiment, thus urement of G in a pure, ballistic environment in Earth orbit allowing multiple measurements that can be optimised follow- that is comparable in accuracy to the terrestrial measurements ing initial experimental results. used in the CODATA definition, using a completely different methodology that is subject to different sources of error. 2 THE SCIENTIFIC REQUIREMENT Even more recent experiments, reported in Reference 3, Gravity is one of the four fundamental forces of nature and G produce G values of 6.674184 × 10−11 and 6.674484 × 10−11 m3 is pivotal to the Newtonian and Einsteinian theories of gravity, kg-1 s-2, with relative standard uncertainties of 11.64 and 11.61

Fig.1 Variations in the measurement of G [2].

172 Vol 72 No.5 May 2019 JBIS THE CONCEPT OF A LOW COST SPACE MISSION to Measure “Big G” parts per million, respectively. While the first measurement agrees with the accepted value of G to within uncertainties, the second result does not – indeed, it is about 45 ppm larger.

3 POTENTIAL SPACE EXPERIMENTS

This paper will briefly review three different classes of experiments carried out in space to measure G. In decreasing order of accuracy, complexity and cost:

3.1 Gravity Train Experiments Fig.2 “Gravity Train” Concept.

There have been several deep space experiments proposed that the technology to be used in a forthcoming ESA mission. aim to measure G by measuring the period of oscillation of a The Laser Interferometer Space Antenna (LISA) will be the small test mass moving within a hollow, much larger and more first space-based gravitational wave observatory. Selected to massive object, of which Reference 4 is the most recent (Fig. 2). be ESA's third large-class mission, it will address the science theme of the Gravitational Universe. LISA will consist of three The setup requires three bodies: a larger layered solid sphere spacecraft separated by 2.5 million km in a triangular forma- with a cylindrical hole through its centre, a much smaller retro- tion, following Earth in its orbit around the Sun. Launch is ex- reflector which will undergo harmonic motion within the hole pected in 2034. and a host spacecraft with laser ranging capabilities to measure round trip light times to the retroreflector but ultimately sepa- The science payload of LISA Pathfinder is shown in Fig. 3. rated a significant distance away from the sphere-retroreflector Although not configured to measure a value of G it is under- apparatus. Measurements of the period of oscillation of the ret- stood that this measurement was attempted at the end of the roreflector in terms of host spacecraft clock time using existing mission, before LP was decommissioned. However, no results technology could give determinations of G nearly three orders are believed to have been published to-date. of magnitude more accurate than current measurements here on Earth. However, significant engineering advances in the re- The inertial sensor subsystem was designed such that a cu- lease mechanism of the apparatus from the host spacecraft will bic test mass located at the centre is free from all external forces likely be necessary. except inertial gravity. The two identical test masses, each one a 46 mm cube composed of gold/platinum alloy, are housed The experiment referenced is proposed to be carried out in in individual evacuated enclosures. The displacement of the deep space – at a distance of 25 AU. The period of oscillation cubes with respect to their housing is measured by capacitive for such an experiment is of the order of hours, and is given by: sensing in three dimensions. These position signals are used in a feedback loop to command proportional micro-propul- (1) sion thrusters to enable the spacecraft to remain centred on the proof mass.

In a preliminary error analysis for a G determination, R is The caging mechanism design and implementation was par- the radius of a near perfect sphere, M is its total mass and T ticularly challenging. The mechanism is required to hold the is the period of the oscillator. The uncertainty in measuring G mass while it is subjected to a very large force (2000 Newton) relies upon the assumption of a sphere outer radius of 4.7 cm during launch, without damaging its surfaces. After launch, and a laser ranging accuracy of 1.1 nm. the test mass must be released at the correct position with high accuracy (60 µm) and with a release speed less than 5 µm per Quantity (Q) Uncertainty (dQ/Q) second in order to allow the electrostatic suspension system to R 7.3 x 10-9 take over the position control. M 5.0 x 10-9 T 1.8 x 10-8 G 6.3 x 10-8

Significant technological advances for the release mechanism from the host are likely to be required to ensure stable initial conditions for the apparatus. An active release mechanism which could reorient the sphere-retroreflector setup if the desired initial conditions were not met would be ideal.

3.2 LISA Pathfinder

An attempt to measure the value of G was reportedly made by the LISA Pathfinder spacecraft (launched in 2015), by measuring the mutual attraction of two test masses using laser metrology, operating at the L1 Lagrange Point of the Sun/Earth system.

The highly successful LISA Pathfinder mission was to test Fig.3 LISA Pathfinder Science Payload [1].

JBIS Vol 72 No.5 May 2019 173 ROGER LONGSTAFF

Fig.4 The residual relative acceleration of the two test masses on ESA's LISA Pathfinder as a function of frequency [1].

The position of the test masses, with respect to the space- and the spacecraft separately to the International Space Station craft or each other, was measured by an interferometric system (ISS) and to integrate the spacecraft there prior to release into that is capable of picometre (10-12 m) precision in the frequen- free flight (thus obviating the complex and heavy caging mech- cy band 10-3 Hz to 10-1 Hz. anisms that were necessary to protect the LISA Pathfinder mission from launch loads). Figure 4 shows the final results of the LISA Pathfinder mission, giving the residual relative acceleration of the two test It is envisaged that the assembled spacecraft could be de- masses. This result substantially exceeded the original science ployed by the ISS CYCLOPS system. The system will utilize goals. NASA's ISS resupply vehicles to launch payloads to the ISS in a controlled pressurized environment in soft stow bags. The 3.3 Free Flying Space Experiments in Low Earth Orbit satellite will then be processed through the ISS pressurized environment by the astronaut crew allowing satellite integration 3.3.1 A Potential Free Flying ISS Launched Mission prior to orbit insertion. Orbit insertion is achieved through use of JAXA's (Japan Aerospace Exploration Agency) Experi- Perhaps the most simple application of a dedicated, free fly- ment Module Robotic Airlock (JEM Airlock), and one of the ing mission to measure G would be to launch the test masses ISS Robotic Arms.

Cyclops will operate from the JEM and take advantage of the airlock's existing slide table, shown in Fig. 5. The launcher will be stowed inside the [space station] for use whenever a satellite is ready to be deployed. Cyclops is placed onto the airlock slide table with the attached satellite and processed through to the external environment. Cyclops, with its attached satellite, is subsequently grasped by the robotic arm and taken to the deployment position. Cyclops then deploys the satellite and is returned to the airlock where it is processed back through and stowed internally for future utilization. The design utilizes the Japanese robotic arm but does have the capability to use the space station's main robotic arm if necessary.

Cyclops interfaces with the JEM Airlock (AL) Slide Table, the ISS Robotic Arms, and the deployable satellites. It is being designed to be able to be utilized for the duration of the ISS mission for the deployment of microsatellites. The platform Fig.5 Illustration of the Cyclops platform configuration. has a size of ~ 127cm L x 61cm W x 7.6cm H, capable of de-

174 Vol 72 No.5 May 2019 JBIS THE CONCEPT OF A LOW COST SPACE MISSION to Measure “Big G” ploying satellites of any geometry up to 100 kg, contingent Technically the two test masses, when released, would be in upon the satellites meeting all Cyclops and ISS safety require- different Earth orbits, causing then to rotate around each other ments. every orbit. However, if the test mass axis of attraction was ar- ranged to be parallel to the ISS velocity vector on release it would An initial concept design for a test mass release and capture be perpendicular to the ISS Earth pointing vector (maintained mechanism is shown in Fig. 6. Note that in this case the dimen- by a constant pitch of 360 degrees in inertial space every orbit). sions of the components have not been optimised, nor a proper In this case the test masses, separated by just a few centimetres, accuracy estimate been made. (More realistic dimensions are would experience almost exactly the same Earth gravitational more likely to resemble the configuration shown in the follow- field, and for the 1 – 3 minute experimental measurement peri- ing section – Newton Pathfinder). od could be assumed to be in a perfect ballistic trajectory.

It is envisaged that the spacecraft containing the science pay- Although not studied in any detail it is thought that such load would be symmetrical in mass with respect to the CoM of an experiment could be designed to produce measurements of the two test masses, and would employ a low thrust electric G that would be comparable in accuracy to the most accurate propulsion, drag makeup thruster to enable free fall conditions terrestrial experiments. for the test masses during measurement periods. A cold gas attitude control system would maintain the test mass attraction 3.3.2 ISS Based Experiment: Newton Pathfinder axis with the velocity vector in order to negate the effects of the Earth’s gravity gradient. Clearly, this would be a complex In a preliminary examination of an ISS based experiment nu- spacecraft, capable of temperature control in periods of sun- merical calculation indicates that a G measurement within 50 light and eclipse and with a communications system to relay ppm is possible with an experiment employing tungsten test telemetry to the Earth, possibly using a store/dump system. masses, each of 5 kg mass and with a centre of mass separation of 10 cm. Such a result would be a valuable scientific meas- A value of G could then be calculated directly from measurement, as it would strengthen the confidence in the CODA- urement of the test masses acceleration: TA definition of G, or alternatively give credence to some of the outlier results (such as measured by the experiment using (2) cold atom interferometry). In practice, at the establishment of the project a more detailed trade-off study would optimise the design by selecting a test mass and CoM separation that Where: would meet the science requirement within the known lim- F – force on test mass (N) its of mechanical design and laser metrology capability. All of M1, M2 – test masses (kg) the required technologies were demonstrated on LISA Path- a1, a2 – acceleration of M1, M2 (m.s-2) finder, and the overall impact of a successful ISS based exper- R – CoM separation (m) iment would be to give impetus to follow-on, highly accurate, G – Gravitational Constant free-flying experiments in space.

TEST MASS CONFIGURATION EXAMPLE

Two copper discs, each 150 mm in diameter, 9.5 mm thick.

Each has a mass of 1.6014 kg.

10.5 Mechanism retracts and allows

20 contactless motion of discs.

Start of separation 10.5 mm Disc position and motion measured C of G of separation 20 mm using laser metrology.

Full closure after 180 seconds.

Mechanism closes and reconfigures discs into starting configuration.

Experiment repeats multiple times in different orbital locations.

Fig.6 Test Mass Release and Capture Mechanism Concept.

JBIS Vol 72 No.5 May 2019 175 ROGER LONGSTAFF

Fig.7 Newton Pathfinder Concept Configuration.

The proposed experimental set up is shown in Fig. 7. The the experiment in the centre of the laboratory where it would basic concept is to measure the self-attraction of test masses in be allowed to “float” (alternatively, the experiment could be po- contactless flight conditions using laser metrology. sitioned using a free flying drone, such as the Astrobee). The astronaut would then exit the module, the laser metrology ap- The experiment container and the test masses would be paratus would be activated and the test masses released by the launched to the ISS separately in a standard cargo transfer bag, simultaneous retraction of the hold and release jaws, and posi- on a routine supply mission, and the experiment would be in- tion measurements telemetered in real time. After each meas- tegrated on-board the ISS by astronauts and closed to be air- urement period the experiment would be reconfigured and tight. The total mass of the experiment is estimated to be about repeated in different orientations in the ISS laboratory (with its 20 kg. This procedure would obviate the need for the complex position and orientation being recorded by CCTV) and at dif- and heavy caging mechanisms that would be required due to ferent orbital locations (such as those with low magnetic flux, launch loads in a fully integrated spacecraft in a dedicated, un- as measured by ISS magnetometers). manned mission (such as were required for Lisa Pathfinder). The experiment would be operated at atmospheric pressure, The experiment would be operated in the pressurised labo- but isolated from drafts within the experiment cage. The “dou- ratory closest to the centre of mass (CoM) of the ISS, in order ble release” procedure (firstly to release the experiment within to minimise the ISS induced gravity gradient between the test the ISS laboratory, and secondly to release the test masses with- masses. (At a known location distant from the ISS CoM the in the experiment cage) would allow free fall conditions for the axis of attraction of the experiment could be aligned perpen- test masses during periods of measurement (initially envisaged dicular to the vector between the experiment axis and the ISS as being 1 – 3 minutes) completely isolated from the (relatively CoM). After the air circulation fans in the laboratory had been noisy) ISS structure microgravity environment. In practice the switched off an astronaut would manually position and release experiment cage would drift within the laboratory in the direc-

176 Vol 72 No.5 May 2019 JBIS THE CONCEPT OF A LOW COST SPACE MISSION to Measure “Big G” tion of the ISS velocity vector, as a consequence of residual air for highly accurate delta displacement measurement by off-the drag on the ISS structure, however, for the short duration of shelf-laser metrology [5], and small enough to ensure contact- measurement this will not affect free-fall conditions within the less conditions with the release & capture mechanisms. cage, as long as it does not drift far enough to make physical contact with the laboratory during measurement. For a 1 micron increase in initial CoM separation to r = 0.100001 m, the displacement after 180 seconds is 4 ACCURACY ESTIMATION FOR NEWTON PATHFINDER 5.424837552977 x 10-4 m, equivalent to an error in G of 20.11 CONCEPT ppm, in agreement with the approximate analytical calculation, while ignoring all other sources of error. 4.1 Simplified Analytical Error Analysis In practice, values of G would be calculated by “curve fit- With an initial (relative) test mass velocity of zero, identical ting” the trajectory data of the test masses as measured by the masses would accelerate toward each other with: laser metrology system.

(3) In summary, both analytical and numerical analysis show that each micron of error in the initial 10 cm CoM separation (for M1 = M2 = M) is equivalent to a 20 ppm of error in G.

Giving a displacement (d) after time t of: This simple, mutual attraction concept has been chosen for a proof of concept experiment as the experiment run time of (4) a few minutes is believed to be compatible with the requirements for contactless conditions. The potentially more accurate oscillation method has a period of order 100 minutes, Which enables a determination of G (for small d,t) of: which is believed to be incompatible with the ISS operating conditions. (5) 5 OTHER SOURCES OF UNCERTAINTY

Therefore, the approximate fractional error in G is: 5.1 Test Mass Centre of Mass Uncertainty

(6) Unpublished communication with Universities participating in the LISA Pathfinder programme suggests that the CoM of the test masses could be measured to an accuracy of approximate- As the uncertainties in t, d and M are judged to be 1 ppm ly 1 micron. The initial distance between the test masses can or better, the uncertainty in r dominates. If, therefore, r can be be measured in a terrestrial Coordinate Measuring Machine determined to 1 micron with a separation of 10 cm, the first (CMM). Reference [6] from the National Physical Laboratory term Equation 6 gives a fractional error in G of (2 x 10-6) / 0.1 gives an accuracy of 250 nm within a measurement volume of = 2 x 10-5 = 20 ppm. 100 mm × 100 mm × 100 mm.

Thus, this simplified analysis shows that each micron uncer- 5.2 Inertial Test mass Uncertainty tainty in CoM separation gives an error in G of 20 ppm. The exact masses of the test masses can be measured much 4.2 Numerical Analysis more accurately than 1 ppm. Reference 7 states that an accuracy of 0.052 ppm can be achieved by a NPL watt balance. A simple computer program using numerical integration was developed to study the self-attraction between the test masses 5.3 Electrostatic Attraction by calculating the force (and hence acceleration) between them and their velocity and position after each integration step. An The effects of electrostatic attraction (caused by charged par- experiment run time of 3 minutes was initially chosen in order ticles penetrating the ISS) can be obviated by “shorting out” to ensure contactless conditions during measurement and suf- the test masses and irradiation with a UV lamp prior to each ficient difference in displacement for accurate scientific anal- experiment run. ysis. An integration step time of one hundredth of a second was initially employed in the analysis. In summary, the baseline 5.4 Magnetic Attraction experimental configuration is: Like terrestrial experiments to measure G, LEO space exper- TM1 = TM2 = 5 kg iments will be influenced by the Earth’s magnetic field. How- ever, the magnetic field will vary with orbital location and ISS CoM initial separation = 10 cm magnetometers measure the field and it can therefore be taken into account when planning experiment run periods. G (CODATA) = 6.674 x 10-11 m3 kg-1 s-2 5.5 Gravity Gradient Experiment run time = 180 seconds If the experiment was carried out at the exact CoM of the ISS Calculation shows that, in the absence of all sources of there would be no gravity gradient from the surrounding struc- error, each (identical) 5 kg test mass would be displaced by ture. In practice the ISS CoM location and mass distribution is 5.425946664237 x 10-4 m after 180 seconds. This is large enough known, and can be compensated for by calculation once the

JBIS Vol 72 No.5 May 2019 177 ROGER LONGSTAFF laboratory that will carry out the experiments has been iden- should enable a programme that could be undertaken within tified. However, the mass of the ISS is 420 tonnes, compared 2 – 3 years and at a cost of < 10 million dollars. This prelim- with terrestrial experiments that are surrounded by thousands inary experiment, while providing a valuable scientific meas- of tonnes of buildings and structure unevenly distributed, lead- urement in its own right, could be looked on as a pathfinder ing to the conclusion that a greater accuracy in a G measure- mission leading to the more accurate free flying experiment ment could be achieved. The Earth’s gravity gradient effect can described above. be deleted by aligning the test mass axis of attraction with the ISS velocity vector. All of the technologies required for the experiment are at a high TRL and the experiment is comparatively simple. Thus 5.6 Laser Metrology Accuracy meeting a delivery deadline within the expected lifetime of the ISS should be easily achievable. It is proposed to verify the de- Miniature, off the shelf devices are available that are accurate to sign and operation of the release mechanism by flying a boiler- nanometre precision, or better [5], and LISA Pathfinder meas- plate version on a parabolic research aircraft. urements were stated to be accurate to picometre precision [1] 7 EDUCATION OUTREACH 5.7 Experiment Timing Accuracy The outreach opportunities given by the experiment are Cheap, miniature, temperature compensated crystal oscillators self-evidently unique and exceptional. Newton Pathfinder is a are commercially available with a timing accuracy of 1 ppm or very simple experiment to grasp in concept and highlights the better. problem of measuring G – which is not generally understood. An astronaut conducting this experiment would provide an 5.8 ISS Microgravity Environment invaluable demonstration of curriculum physics and it would also be an excellent introduction to microgravity research and As explained in section 3.3.2, the test masses during measure- its associated technologies. ment periods are doubly isolated from the ISS, and the atmosphere within the experiment cage is isolated from the labora- 8 HIGHLY SPECULATIVE SCIENTIFIC INVESTIGATION tory atmosphere. Air drag will cause the experiment cage to move about 1 cm in the direction of the ISS velocity vector dur- Conventional wisdom defines G as a temporally and spatially ing the measurement period, and will not come into physical invariant scalar quantity. It seems certain that G is temporally contact with anything within the laboratory. invariant over centennial timescales, as planetary ephemer- is observations give no evidence to the contrary and current Thus, in a preliminary investigation it appears that the dom- theories do not suggest that G has altered over the observable inant contribution to uncertainty in a G determination by this lifetime of the universe. However, observations of the Cosmic method is the initial error in Centre of Mass Separation. Microwave Background have shown a temperature variation of the order of 100 ppm over the celestial sphere, which could 6 THE COST OF SPACE EXPERIMENTS lead to speculation of a similar variation in G. As the experiment described could obtain measurements with relatively The costs of the three classes of experiments described have stable celestial coordinates it could be possible to search for not been formally estimated, however, it seems likely that there any repeatable pattern in alignment with the celestial sphere. is at least an order of magnitude difference with each level of It is not believed that such a hypothesis has been investigated system complexity. before. However unlikely, any indication of a spatial depend- ence in G would have a profound effect on theoretical physics The most costly experiment is clearly the deep space mis- and cosmology. sion to measure G employing the classic, gravity train method, as described in [1]. This involves a multi-year spaceflight to a 9 FUTURE WORK location in the solar system stated to be >25 AU. It is inconceiv- able that such a mission could be undertaken at an overall cost This paper reports on a concept study that has not found any of less than hundreds of millions of dollars when considering obvious “showstoppers” to a relatively inexpensive mission the costs of design, test, launch and flight operations. to measure G in space. It could be regarded as a pre-phase A study, in ESA terminology. Clearly, if the project were to ad- The free flying experiment briefly envisaged should cost vance it would require a proper Phase A feasibility study, in- roughly an order of magnitude less – very approximately 10s cluding specialists in gravitational physics, laser metrology, of millions of dollars. The cost saving comes from a (relative- precision manufacturing and ISS operations. ly) cheap launch on a routine ISS supply mission to low Earth orbit, the use of technology flight proven by the LISA Path- 10 CONCLUSIONS finder mission, and astronaut intervention for final assembly (obviating heavy, complex and costly caging mechanisms for A preliminary investigation has found that a relatively simple the test masses). and inexpensive experiment, conducted on-board the Inter- national Space Station and using existing, flight proven tech- The Newton Pathfinder experiment would be the least cost- nology, could help to resolve a current problem in experimen- ly of the three options as it would benefit from all of the above tal physics – the determination of the Universal Gravitational advantages shown above as well as operation by astronauts in Constant – G. The value of such an experiment would be that atmospheric conditions on-board the ISS. The use of fast pro- the measurement would be made using a completely new and totyping in university laboratories, off the shelf instruments different method, with different sources of error from those and equipment, and the measurement capabilities available at applying to terrestrial measurements that do not agree with government facilities (e.g. the National Physical Laboratory) each other.

178 Vol 72 No.5 May 2019 JBIS THE CONCEPT OF A LOW COST SPACE MISSION to Measure “Big G”

Acknowledgements modern language. Thanks are also due to the anonymous reviewers whose comments have (hopefully) improved this pa- The author would like to thank Mr. Mark Hempsell for proper. Any errors in this analysis are those of the author, and the ducing CAD drawings of experiment designs, and Mr. Peter author alone. Meggitt for translating the author’s computer program into a

REFERENCES 1. M .Armano et al., “Beyond the Required LISA Free-Fall Performance, 5. UK Space Agency Website. https://www.gov.uk/government/uploads/ Physical Review Letters (2018) system/uploads/attachment_data/file/573242/An_optical_fibre_sensor_ 2. National Science Foundation Website, https://www.nist.gov/news- to_measure_nanometre_changes_in_position.pdf events/events/2016/07/nsf-ideas-laboratory-nist-measuring-big-g- 6. National Physical Laboratory Website, High Accuracy Micro challenge Coordinate Measuring Machine, https://www.npl.co.uk/products- 3. Qing Li, “Measurements of the gravitational constant using two services/dimensional/high-accuracy-coordinate-measuring-machine independent methods”, Nature 560, 582-588 (2018) 7. L. Chao et al, “The Design of the new NIST-4 Watt Balance”, Conference 4. M.R. Feldman et al, “Deep Space Experiment to Measure G”, Classical on Precision Electromagnetic Measurements, August 2014 and Quantum Gravity 33 (12), May 2016

Received 25 June 2019 Approved 12 July 2019

JBIS Vol 72 No.5 May 2019 179 CALL FOR PAPERS

Putting Astronauts in Impossible Locations A one day technical symposium

9:00 a.m - Wednesday 27 November 2019

BIS HQ, 27/29 South Lambeth Road

While the human exploration of the Moon and Mars has been extensively examined, serious technical consideration of the rest of the solar system has been largely ignored. This symposium is designed to explore the limits of where human exploration can go in the solar system and how to overcome the challenges involved. The symposium is open to papers on the transportation requirements, the practicalities of habitation in extreme environments and any other aspects of a solar system wide civilisation. Submissions should be on the basis that there will be a completed paper delivered before the symposium as well as giving a presentation on the day. All papers will be considered for publication in the Journal of the British Interplanetary Society.

Proposed papers should be described in an abstract of no more than 400 words, and submitted to the Society at [email protected]

Submission deadline 31 July 2019

Journal of the British Interplanetary Society

VOLUME 72 NO.5 MAY 2019

www.bis-space.com

ISSN 0007-084X PUBLICATION DATE: 26 JULY 2019